diff --git a/.gitattributes b/.gitattributes index a6344aac8c09253b3b630fb776ae94478aa0275b..52373fe24473b1aa44333d318f578ae6bf04b49b 100644 --- a/.gitattributes +++ b/.gitattributes @@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text *.zip filter=lfs diff=lfs merge=lfs -text *.zst filter=lfs diff=lfs merge=lfs -text *tfevents* filter=lfs diff=lfs merge=lfs -text +tokenizer.json filter=lfs diff=lfs merge=lfs -text diff --git a/chat_template.jinja b/chat_template.jinja new file mode 100644 index 0000000000000000000000000000000000000000..33089ace1be88f22a10fe861ad49718d5d886090 --- /dev/null +++ b/chat_template.jinja @@ -0,0 +1,109 @@ +{{- bos_token }} +{%- if custom_tools is defined %} + {%- set tools = custom_tools %} +{%- endif %} +{%- if not tools_in_user_message is defined %} + {%- set tools_in_user_message = true %} +{%- endif %} +{%- if not date_string is defined %} + {%- set date_string = "26 Jul 2024" %} +{%- endif %} +{%- if not tools is defined %} + {%- set tools = none %} +{%- endif %} + +{#- This block extracts the system message, so we can slot it into the right place. #} +{%- if messages[0]['role'] == 'system' %} + {%- set system_message = messages[0]['content']|trim %} + {%- set messages = messages[1:] %} +{%- else %} + {%- set system_message = "" %} +{%- endif %} + +{#- System message + builtin tools #} +{{- "<|start_header_id|>system<|end_header_id|>\n\n" }} +{%- if builtin_tools is defined or tools is not none %} + {{- "Environment: ipython\n" }} +{%- endif %} +{%- if builtin_tools is defined %} + {{- "Tools: " + builtin_tools | reject('equalto', 'code_interpreter') | join(", ") + "\n\n"}} +{%- endif %} +{{- "Cutting Knowledge Date: December 2023\n" }} +{{- "Today Date: " + date_string + "\n\n" }} +{%- if tools is not none and not tools_in_user_message %} + {{- "You have access to the following functions. To call a function, please respond with JSON for a function call." }} + {{- 'Respond in the format {"name": function name, "parameters": dictionary of argument name and its value}.' }} + {{- "Do not use variables.\n\n" }} + {%- for t in tools %} + {{- t | tojson(indent=4) }} + {{- "\n\n" }} + {%- endfor %} +{%- endif %} +{{- system_message }} +{{- "<|eot_id|>" }} + +{#- Custom tools are passed in a user message with some extra guidance #} +{%- if tools_in_user_message and not tools is none %} + {#- Extract the first user message so we can plug it in here #} + {%- if messages | length != 0 %} + {%- set first_user_message = messages[0]['content']|trim %} + {%- set messages = messages[1:] %} + {%- else %} + {{- raise_exception("Cannot put tools in the first user message when there's no first user message!") }} +{%- endif %} + {{- '<|start_header_id|>user<|end_header_id|>\n\n' -}} + {{- "Given the following functions, please respond with a JSON for a function call " }} + {{- "with its proper arguments that best answers the given prompt.\n\n" }} + {{- 'Respond in the format {"name": function name, "parameters": dictionary of argument name and its value}.' }} + {{- "Do not use variables.\n\n" }} + {%- for t in tools %} + {{- t | tojson(indent=4) }} + {{- "\n\n" }} + {%- endfor %} + {{- first_user_message + "<|eot_id|>"}} +{%- endif %} + +{%- for message in messages %} + {%- if not (message.role == 'ipython' or message.role == 'tool' or 'tool_calls' in message) %} + {{- '<|start_header_id|>' + message['role'] + '<|end_header_id|>\n\n'+ message['content'] | trim + '<|eot_id|>' }} + {%- elif 'tool_calls' in message %} + {%- if not message.tool_calls|length == 1 %} + {{- raise_exception("This model only supports single tool-calls at once!") }} + {%- endif %} + {%- set tool_call = message.tool_calls[0].function %} + {%- if builtin_tools is defined and tool_call.name in builtin_tools %} + {{- '<|start_header_id|>assistant<|end_header_id|>\n\n' -}} + {{- "<|python_tag|>" + tool_call.name + ".call(" }} + {%- for arg_name, arg_val in tool_call.arguments | items %} + {{- arg_name + '="' + arg_val + '"' }} + {%- if not loop.last %} + {{- ", " }} + {%- endif %} + {%- endfor %} + {{- ")" }} + {%- else %} + {{- '<|start_header_id|>assistant<|end_header_id|>\n\n' -}} + {{- '{"name": "' + tool_call.name + '", ' }} + {{- '"parameters": ' }} + {{- tool_call.arguments | tojson }} + {{- "}" }} + {%- endif %} + {%- if builtin_tools is defined %} + {#- This means we're in ipython mode #} + {{- "<|eom_id|>" }} + {%- else %} + {{- "<|eot_id|>" }} + {%- endif %} + {%- elif message.role == "tool" or message.role == "ipython" %} + {{- "<|start_header_id|>ipython<|end_header_id|>\n\n" }} + {%- if message.content is mapping or message.content is iterable %} + {{- message.content | tojson }} + {%- else %} + {{- message.content }} + {%- endif %} + {{- "<|eot_id|>" }} + {%- endif %} +{%- endfor %} +{%- if add_generation_prompt %} + {{- '<|start_header_id|>assistant<|end_header_id|>\n\n' }} +{%- endif %} diff --git a/config.json b/config.json new file mode 100644 index 0000000000000000000000000000000000000000..c1c2d6a4f2a3f026f3b25b8a645fd45937dbd80f --- /dev/null +++ b/config.json @@ -0,0 +1,82 @@ +{ + "architectures": [ + "LlamaForCausalLM" + ], + "attention_bias": false, + "attention_dropout": 0.0, + "bos_token_id": 128000, + "eos_token_id": [ + 128001, + 128008, + 128009 + ], + "head_dim": 128, + "hidden_act": "silu", + "hidden_size": 16384, + "initializer_range": 0.02, + "intermediate_size": 53248, + "max_position_embeddings": 131072, + "mlp_bias": false, + "model_type": "llama", + "num_attention_heads": 128, + "num_hidden_layers": 126, + "num_key_value_heads": 8, + "pad_token_id": 128009, + "pretraining_tp": 1, + "quantization_config": { + "config_groups": { + "group_0": { + "input_activations": { + "actorder": null, + "block_structure": null, + "dynamic": true, + "group_size": null, + "num_bits": 8, + "observer": null, + "observer_kwargs": {}, + "strategy": "token", + "symmetric": true, + "type": "int" + }, + "output_activations": null, + "targets": [ + "Linear" + ], + "weights": { + "actorder": null, + "block_structure": null, + "dynamic": false, + "group_size": null, + "num_bits": 8, + "observer": "minmax", + "observer_kwargs": {}, + "strategy": "channel", + "symmetric": true, + "type": "int" + } + } + }, + "format": "int-quantized", + "global_compression_ratio": null, + "ignore": [ + "lm_head" + ], + "kv_cache_scheme": null, + "quant_method": "compressed-tensors", + "quantization_status": "compressed" + }, + "rms_norm_eps": 1e-05, + "rope_scaling": { + "factor": 8.0, + "high_freq_factor": 4.0, + "low_freq_factor": 1.0, + "original_max_position_embeddings": 8192, + "rope_type": "llama3" + }, + "rope_theta": 500000.0, + "tie_word_embeddings": false, + "torch_dtype": "bfloat16", + "transformers_version": "4.52.3", + "use_cache": false, + "vocab_size": 128256 +} \ No newline at end of file diff --git a/generation_config.json b/generation_config.json new file mode 100644 index 0000000000000000000000000000000000000000..329d2aad7a332f09a4a3a035e5eb1f37a8ae84a2 --- /dev/null +++ b/generation_config.json @@ -0,0 +1,12 @@ +{ + "bos_token_id": 128000, + "do_sample": true, + "eos_token_id": [ + 128001, + 128008, + 128009 + ], + "temperature": 0.6, + "top_p": 0.9, + "transformers_version": "4.52.3" +} diff --git a/model-00001-of-00086.safetensors b/model-00001-of-00086.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..29e745b98e37f303203f69847e860c2b274f1fa3 --- /dev/null +++ b/model-00001-of-00086.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:1c32de0435cfa577817c38cf6dda2617a126055c6c84e0f3ea0da561683085e6 +size 4773188672 diff --git a/model-00002-of-00086.safetensors b/model-00002-of-00086.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..142a4155ea00d8d74d59ee93cd2ab71e926af9be --- /dev/null +++ b/model-00002-of-00086.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c7023bbe0146e98cc8c2da9708b173b17ee7333b4d302bdeca5ea3bba314440c +size 4933097696 diff --git a/model-00003-of-00086.safetensors b/model-00003-of-00086.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..32f2027b19237d2253ea06791e193671088cd4ed --- /dev/null +++ b/model-00003-of-00086.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:5ffc7a0747bbe4676446c104c29a09cf49a8e59430c20e2030f5d38d4202a219 +size 4631063664 diff --git a/model-00004-of-00086.safetensors b/model-00004-of-00086.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..f518945b0fb7c10f36c21b8ef0336fc2b7cc7515 --- /dev/null +++ b/model-00004-of-00086.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:bbf5a5b4b167ac84390e0cebadd1e11ea3143278a6f9e826414f044ce72911ee +size 4933097696 diff --git a/model-00005-of-00086.safetensors b/model-00005-of-00086.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..f25875d567dc04bb52c0a2bc7358c3ea674fa8ff --- /dev/null +++ b/model-00005-of-00086.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f6ef0790ed9e5d3c7fd24737f0c15bf276d643207f91a24856924b95e11a7e16 +size 4631063664 diff --git a/model-00006-of-00086.safetensors b/model-00006-of-00086.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..a388ba06951d0686be22d7ce25b2365bb2929251 --- /dev/null +++ b/model-00006-of-00086.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:83937428d21efa24b0f1a98404af7925193e22c00afdaeae6ecb14cd4dd0f4fd +size 4933097696 diff --git a/model-00007-of-00086.safetensors b/model-00007-of-00086.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..d6806f3341d78046f3acad02810d3370d8c41e4a --- /dev/null +++ b/model-00007-of-00086.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:bfcfb13be51b8c51d1164802ee7a36e75441111517f0cf9963dec7ed33dba443 +size 4631063664 diff --git a/model-00008-of-00086.safetensors b/model-00008-of-00086.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..5cab4b249f8b0cab35c0fa136d53270d06f6167e --- /dev/null +++ b/model-00008-of-00086.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7899f83b7da401ea89f84c3bc4aa57365cd2ddb4ca1255c34313d405cc302018 +size 4933097712 diff --git a/model-00009-of-00086.safetensors b/model-00009-of-00086.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..d53badc11429dd60f7a809b69a7446e36485db30 --- /dev/null +++ b/model-00009-of-00086.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:60ed66b88ebb8d1e0503975dfa83ce853ee7f6415bc03d105f10f4c383b24851 +size 4631063696 diff --git a/model-00010-of-00086.safetensors b/model-00010-of-00086.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..91a190d0f3391e930960b6ba248c075fc83e6687 --- /dev/null +++ b/model-00010-of-00086.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0000f1848dc781d67e20b4931a6e9a21271968179a2b3e5eec6d7988efb5f3a0 +size 4933097720 diff --git a/model-00011-of-00086.safetensors b/model-00011-of-00086.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..672a6204cb8f42b06a1586a14d2b6eb4e65eec73 --- /dev/null +++ b/model-00011-of-00086.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:031ae055eb2fd85bf38825f7e205e7c7f80d4803dd53c3f65e3b7ed22fb791d9 +size 4631063696 diff --git a/model-00012-of-00086.safetensors b/model-00012-of-00086.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..0c9a28707b42029602a0aeb53a79518e8b316282 --- /dev/null +++ b/model-00012-of-00086.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:66e9ee110b648d87de80e43a3d153f9ada37fddf80cd659ba7be9f28e971b424 +size 4933097720 diff --git a/model-00013-of-00086.safetensors b/model-00013-of-00086.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..87b6ecb3b214cfbb2284549de17568f80a9aaf6e --- /dev/null +++ b/model-00013-of-00086.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:5e39056b74473e11846b96f307ec14e6f2b2d528545f000d38fb2f7a5df6be9d +size 4631063696 diff --git a/model-00014-of-00086.safetensors b/model-00014-of-00086.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..19fc70fd8415b4af5b28d8324e8fe5d0a8c22f3f --- /dev/null +++ b/model-00014-of-00086.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:9a6edf3d90197f2fd4be8b3689d8d2337a41430cf76a7c678c8c2bfb2247a6c2 +size 4933097720 diff --git a/model-00015-of-00086.safetensors b/model-00015-of-00086.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..3c165f1514649080922d8bc9c14bd50f1dbedf01 --- /dev/null +++ b/model-00015-of-00086.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7ec340542c6e2197dc1bc0e61e75da41b3dc4afce6b8283e2b4a91af5c3d22c7 +size 4631063696 diff --git a/model-00016-of-00086.safetensors b/model-00016-of-00086.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..7ebf1d83a024a41a0b6b5614804d5bb9b7c59efc --- /dev/null +++ b/model-00016-of-00086.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c7051e51beb70abc633c1284d5b5a1e523eddfa90dd73548dc3e7a0b128a6ae4 +size 4933097720 diff --git a/model-00017-of-00086.safetensors b/model-00017-of-00086.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..dd4625aefef06c24b55e9defa3278a55a6ecd718 --- /dev/null +++ b/model-00017-of-00086.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8af561633fa219db21052a0eccd15732c365929200ff9b84c645a771df4909c2 +size 4631063696 diff --git a/model-00018-of-00086.safetensors b/model-00018-of-00086.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..3c12dc75b7510c3ff9d288e4f3aa94011fd2012b --- /dev/null +++ b/model-00018-of-00086.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:99c96e2f49a8069e32e94dc41ef8fdb0412641afcaf6b165c422ba2d4d523277 +size 4933097720 diff --git a/model-00019-of-00086.safetensors b/model-00019-of-00086.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..abcfa63dbf63198c5548c81eb6e7a0704fec3473 --- /dev/null +++ b/model-00019-of-00086.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c5a6547103dc271cdd3fbb8f9232e2afcf0c65225755b9d0a1433a337f384d8c +size 4631063696 diff --git a/model-00020-of-00086.safetensors b/model-00020-of-00086.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..7b03f3db902c1f99869782991379058b7925f83e --- /dev/null +++ b/model-00020-of-00086.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:32a09b1d8ed45c123ccba96ce4d164a3b39473f2d29b5543fee550ada10b3cd8 +size 4933097720 diff --git a/model-00021-of-00086.safetensors b/model-00021-of-00086.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..72e7e7ceca6a84edf76fbd69e556e3294b1f59b9 --- /dev/null +++ b/model-00021-of-00086.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:37fd8a0196b8a9f0f3a53890b738e1f4da6cebe772ea15279fc396a766fec127 +size 4631063696 diff --git a/model-00022-of-00086.safetensors b/model-00022-of-00086.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..42c03cd99941bdb64b10dfb23cfed7da6be018ec --- /dev/null +++ b/model-00022-of-00086.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:5a9cd168e19da2772a11d8a609445802c301993e7626d24a5f57195bbbe5faf1 +size 4933097720 diff --git a/model-00023-of-00086.safetensors b/model-00023-of-00086.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..f4d801b8603a7186526a157acbedf735cfdb68ae --- /dev/null +++ b/model-00023-of-00086.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a347f6340b2643e564d2c2b41a28c073cb554f18d0511f11a675fb8827bb3e9d +size 4631063696 diff --git a/model-00024-of-00086.safetensors b/model-00024-of-00086.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..4cfbe9dc756d22a5b47aa04d283fa229df8588bd --- /dev/null +++ b/model-00024-of-00086.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:215e1e72fb33681a7aca3af321b03808a521301837192258b4095c1909bb5b1e +size 4933097720 diff --git a/model-00025-of-00086.safetensors b/model-00025-of-00086.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..0edda19deee0665780bd8198550dfeb7bb50735e --- /dev/null +++ b/model-00025-of-00086.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:1ec0f54170817991685755b8fe6146912a5413fc06371672b6884ff74dc9e874 +size 4631063696 diff --git a/model-00026-of-00086.safetensors b/model-00026-of-00086.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..649136822756742865c03e7a121f6aab345e6477 --- /dev/null +++ b/model-00026-of-00086.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:444cd0660fd91c7789251e1431514b0e9d368e6e64bbcc96cf26084328a69120 +size 4933097720 diff --git a/model-00027-of-00086.safetensors b/model-00027-of-00086.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..e3f9c8e78b92708881d734808d92e246c03b4618 --- /dev/null +++ b/model-00027-of-00086.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:dd8736c3f615c6df8c1ee62a1ba49f6ff68a3aa49d419c2c56ef907fde9aa2a0 +size 4631063696 diff --git a/model-00028-of-00086.safetensors b/model-00028-of-00086.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..e9801dee294820171f981aa6c0eab887471f055e --- /dev/null +++ b/model-00028-of-00086.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:23a1af7212f28d8de61175175476228f334340d9b51eb6f2cf526f790bd93fee +size 4933097720 diff --git a/model-00029-of-00086.safetensors b/model-00029-of-00086.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..5698cfe0e0a353cbb227442ce9b5a91c879e1f3d --- /dev/null +++ b/model-00029-of-00086.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0bfaa4bee8f3c9335761ff33d76ddbccdf99780f826f99c4f057577276b83a09 +size 4631063696 diff --git a/model-00030-of-00086.safetensors b/model-00030-of-00086.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..d29a6d5f84e3a3ae299ae3d2dde6e71cc74e1441 --- /dev/null +++ b/model-00030-of-00086.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a14a35bfcc986cf2c9b03bc3752d44eda8425efb5ddd98201f358212e1873724 +size 4933097720 diff --git a/model-00031-of-00086.safetensors b/model-00031-of-00086.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..13f09a6868865c60be07ff95cfcf0b1d325feb43 --- /dev/null +++ b/model-00031-of-00086.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:52e135f00386f90d006adfe3d4ce6d0aeca7a3534d757bdbccaea5be54e5a6bd +size 4631063696 diff --git a/model-00032-of-00086.safetensors b/model-00032-of-00086.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..18baa448fd37a72fdd48ed4fee8f1d4891998b4d --- /dev/null +++ b/model-00032-of-00086.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:44fc3cff0edf13c0d34da9656b641b75b8ed9fb155420c32b9df9a2780e32631 +size 4933097720 diff --git a/model-00033-of-00086.safetensors b/model-00033-of-00086.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..4bf1845aca405305812d0f3ac29acfb6262ac4f7 --- /dev/null +++ b/model-00033-of-00086.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:02ecb0bcc796d13ed0789961222e79dced2b4c2ddbf76958a166e194c4b5d223 +size 4631063696 diff --git a/model-00034-of-00086.safetensors b/model-00034-of-00086.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..fddfd5ef731c1879b8ba845db89b8fb1157269d6 --- /dev/null +++ b/model-00034-of-00086.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:50ca948e86e9ce49dc60809319bc903dea206f42956ced4d82c5b082ba386cb8 +size 4933097720 diff --git a/model-00035-of-00086.safetensors b/model-00035-of-00086.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..34579a6bc3e44c184f4b1a2c471c760c33f12ddd --- /dev/null +++ b/model-00035-of-00086.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:af91aea60976d8207e0602da8bffe1e1673805ed796705c4792ea67cc5c4a969 +size 4631063696 diff --git a/model-00036-of-00086.safetensors b/model-00036-of-00086.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..8914ea8cb87142747df7b78bc33e6db5e4b12135 --- /dev/null +++ b/model-00036-of-00086.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:59ac6dd428b628ccd419dd40cd7412cba87bd3133ee489cc7df26b463f091055 +size 4933097720 diff --git a/model-00037-of-00086.safetensors b/model-00037-of-00086.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..afc1816a5e1622e4730e542eb0f5c7dd4d74fd68 --- /dev/null +++ b/model-00037-of-00086.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:764d6e58005a3e0f029cf60930d92c96b13016564195d67fe0206bd755d6a16b +size 4631063696 diff --git a/model-00038-of-00086.safetensors b/model-00038-of-00086.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..e30392b75eae335ebccc27cf60162d2ffb7bd4ed --- /dev/null +++ b/model-00038-of-00086.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f6e2164be92d3789b34139eeb860f5aee08ed2f001262cb328c6c3a17b113cb1 +size 4933097720 diff --git a/model-00039-of-00086.safetensors b/model-00039-of-00086.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..26068f4edf078c8ae2dc5bcba67f82cfad8c34a0 --- /dev/null +++ b/model-00039-of-00086.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:4b8e5f08400bf2f45beefd7707cc980d89e15e0c182ce17cd5f5a649dc578eea +size 4631063696 diff --git a/model-00040-of-00086.safetensors b/model-00040-of-00086.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..ee6f007aa35427602f088a520be35170dd83774e --- /dev/null +++ b/model-00040-of-00086.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:40312281402fef32bd1850efe79831862c9921c269705a5236779f0127160fa7 +size 4933097720 diff --git a/model-00041-of-00086.safetensors b/model-00041-of-00086.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..c2e5a31bfa78a561adb5c45dcd593817949fc79c --- /dev/null +++ b/model-00041-of-00086.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:37dd9ff08711ebe2af6c039b087e9b0bee83ad82876988bbf97bf3598bf562f0 +size 4631063696 diff --git a/model-00042-of-00086.safetensors b/model-00042-of-00086.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..b2fac03373b0d08b1d817a9f23dfa72bd6b40776 --- /dev/null +++ b/model-00042-of-00086.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:af6df0d64644996aba91f30058c75403733236304fcfb147adf1754486caa891 +size 4933097720 diff --git a/model-00043-of-00086.safetensors b/model-00043-of-00086.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..d0f58337006ed559c163a5eca48c2675a7e2e405 --- /dev/null +++ b/model-00043-of-00086.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:dad670ff4bacd3d204769c62ae925864f47b990aa64e08393cd49877a7e3b8f4 +size 4631063696 diff --git a/model-00044-of-00086.safetensors b/model-00044-of-00086.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..be6b9117d5fa4d2d119b263f1a74206570a2e1a3 --- /dev/null +++ b/model-00044-of-00086.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:73be589f58dd7410dbe4e69d54af50be85f69d6c27319b587f6c4aa8f7830c83 +size 4933097720 diff --git a/model-00045-of-00086.safetensors b/model-00045-of-00086.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..0520d89b4d6764af400d153d521421e32e60a41f --- /dev/null +++ b/model-00045-of-00086.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:fb21bf7523d6dee8d64bbe158972c750f8060c975bfc16d29afc7d600fa01d98 +size 4631063696 diff --git a/model-00046-of-00086.safetensors b/model-00046-of-00086.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..e2218a5bef9cac5acff73240026ff55d0a6b8f25 --- /dev/null +++ b/model-00046-of-00086.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3aef854735b292cbd8f270ba0dd884ae543299f0250eabd3f36df8714f210849 +size 4933097720 diff --git a/model-00047-of-00086.safetensors b/model-00047-of-00086.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..1a60b4fe5cd2f3dc20179d3ba00f334de3fbc475 --- /dev/null +++ b/model-00047-of-00086.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:16181f900b08c4e7a44e75dcd8fff3d5475d2d18737a118e1f9417b1f0cf6037 +size 4631063696 diff --git a/model-00048-of-00086.safetensors b/model-00048-of-00086.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..b5fc16ea2cca104ef78b41e0829f01b5bd5dc3b9 --- /dev/null +++ b/model-00048-of-00086.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:96a6469f77204bc27bf76484bfdffddf2b5db50b913798470b089696ee992c18 +size 4933097720 diff --git a/model-00049-of-00086.safetensors b/model-00049-of-00086.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..6c81cf3aec6b4cfaf7453e72e3fee9c92b396153 --- /dev/null +++ b/model-00049-of-00086.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:97a6d2f4e48b130f26b7b1dfddb45e504e2e227a2905a78e91ea1b5dc5318136 +size 4631063696 diff --git a/model-00050-of-00086.safetensors b/model-00050-of-00086.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..c4809ef94cf22ec964a262e84f00cd6020a901bd --- /dev/null +++ b/model-00050-of-00086.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8ae6e5ff6e23a28d30d594753f755c9978c96ca9104fc8ea3fc20e1358f5bf61 +size 4933097720 diff --git a/model-00051-of-00086.safetensors b/model-00051-of-00086.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..158169e7105bf11078702b99eaf25ca1dce25f49 --- /dev/null +++ b/model-00051-of-00086.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:49c9343ef5cac0ed3b5d76c65c0cb74585e8c1300232f307eb8245aa3ac3bd2f +size 4631063696 diff --git a/model-00052-of-00086.safetensors b/model-00052-of-00086.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..1760f9409ed7d8a6503ce10255533d76c8075a2b --- /dev/null +++ b/model-00052-of-00086.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e4e8efd9542251af8920f3fc1328d82ef8c16dfe563b1650d9ab5c8539f459d6 +size 4933097720 diff --git a/model-00053-of-00086.safetensors b/model-00053-of-00086.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..e969ff6052b9f7bee294c9b7c3e23893a9cc8bbe --- /dev/null +++ b/model-00053-of-00086.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:1c4b5cca2f0bd3722346c3f4cd05fd73ae195f587e6e1b423dbe8eb24952c198 +size 4631063696 diff --git a/model-00054-of-00086.safetensors b/model-00054-of-00086.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..9d6118ff8e27ce37949cf9223803df0efdfe4319 --- /dev/null +++ b/model-00054-of-00086.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7a3c09b93861107a3467504a4aa23d32141d99b29c4d1f229397683978cfec84 +size 4933097720 diff --git a/model-00055-of-00086.safetensors b/model-00055-of-00086.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..c0d6d9556f10cd0d261f4ec33f03c3a6c845f7d4 --- /dev/null +++ b/model-00055-of-00086.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:1a9655a640c282c887f6b34282ab758d9c93c19cd928e4536b72441fdc480370 +size 4631063696 diff --git a/model-00056-of-00086.safetensors b/model-00056-of-00086.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..ce4eaf7db6ac25c83bbadeffca8900fc6ad91e26 --- /dev/null +++ b/model-00056-of-00086.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6e25b4b798b870aad3fc153b5458c82cefbfa58cef3ea22738fa6d8e322d0125 +size 4933097720 diff --git a/model-00057-of-00086.safetensors b/model-00057-of-00086.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..8d5af74a62689f6323c37905de279b005427c806 --- /dev/null +++ b/model-00057-of-00086.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:cba40361011a7eb586c28858fcf8bf289174f35783c055c1579d5b2cd1829dcb +size 4631063696 diff --git a/model-00058-of-00086.safetensors b/model-00058-of-00086.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..5884beda1d42c92530d7cd4c81fbe63431550cba --- /dev/null +++ b/model-00058-of-00086.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:eb6fec9074c0ee99ed000da400d24023aab9c91772cd1558ef073dcf9bc94078 +size 4933097720 diff --git a/model-00059-of-00086.safetensors b/model-00059-of-00086.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..481c00a100316d07dd530419a6fdf3b8a8feda99 --- /dev/null +++ b/model-00059-of-00086.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e342349cbd36c2fc55d45b61d410f9d69b10077cc0b6d9bb7c9d6df2865294c7 +size 4631063696 diff --git a/model-00060-of-00086.safetensors b/model-00060-of-00086.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..12a323bfbd026da29d12efe5af669ee94d4939d6 --- /dev/null +++ b/model-00060-of-00086.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f757445f89133093036077c745cbee3f9b7a678f5f06c8fafebc6ffe877c51fe +size 4933097720 diff --git a/model-00061-of-00086.safetensors b/model-00061-of-00086.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..51d7a6eb808d7f74f55a1d215819101d0287163a --- /dev/null +++ b/model-00061-of-00086.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:4aeec3a759792b41a077064a4d139be857fa33bbc2418c4c24df276362b8ce4c +size 4631063696 diff --git a/model-00062-of-00086.safetensors b/model-00062-of-00086.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..52cce71ccc9df41cd76d405f75f4cbae16e0077a --- /dev/null +++ b/model-00062-of-00086.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:681c05c6bf426418b9334cfd46d502db7e7df37ed19e3961d47ecea808b26f46 +size 4933097720 diff --git a/model-00063-of-00086.safetensors b/model-00063-of-00086.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..0429fa7d84d5d2f1a4754bf56bda79698ae1afd0 --- /dev/null +++ b/model-00063-of-00086.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:9bbb3261cf5ca2108c7c6f5a6957eead76ec78a1b62f45004474bf6002923b37 +size 4631063696 diff --git a/model-00064-of-00086.safetensors b/model-00064-of-00086.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..93f6ff16d2febceb65669eed465d46fe079ae870 --- /dev/null +++ b/model-00064-of-00086.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:824baef8f482de298efea9611f7d136b10fc0e5e34a28a7413ba357c9ea77ed7 +size 4933097720 diff --git a/model-00065-of-00086.safetensors b/model-00065-of-00086.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..72b2bf05aff4bd95ca0c9d13ecae2834d52bf390 --- /dev/null +++ b/model-00065-of-00086.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e1e413a2273314286844c636033df56a99b8559f9faf565d33928a1709bf000c +size 4631063696 diff --git a/model-00066-of-00086.safetensors b/model-00066-of-00086.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..61b14aaae05a4697e7b26da03e7f7ec54ff97700 --- /dev/null +++ b/model-00066-of-00086.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:b7a934520777d00fa7b83fefd7098402e63b157ea2d71b627b97265bbbffcaa9 +size 4933097720 diff --git a/model-00067-of-00086.safetensors b/model-00067-of-00086.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..ef6ab46377c3cc424ddcb548aa29a675fd812545 --- /dev/null +++ b/model-00067-of-00086.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:46c399478353c1c239dcd2cca2bf8748d802c2a34d226c309b34f1676e8af4f1 +size 4631063696 diff --git a/model-00068-of-00086.safetensors b/model-00068-of-00086.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..d1fb4ec80e23ed9fc8208516cd6210715ffdc8a3 --- /dev/null +++ b/model-00068-of-00086.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:cca2f49feaaa75b787e7d62110204df3d16751fc602f0505d41404e34a75a6ad +size 4933097736 diff --git a/model-00069-of-00086.safetensors b/model-00069-of-00086.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..4868b32e4c87aa489d108edd217882ffbfd5bd70 --- /dev/null +++ b/model-00069-of-00086.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8865a7c26c24b32c073fa7848410734b04fdbd3d43ec7ad0da78934b7916249f +size 4631063720 diff --git a/model-00070-of-00086.safetensors b/model-00070-of-00086.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..87d9d5903f7f8161d753dec0d8278dfe98886aa7 --- /dev/null +++ b/model-00070-of-00086.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7505a2b86863cf2534cbb6b71e033f6c05f6aaed25350fd95ef7124c58757d93 +size 4933097736 diff --git a/model-00071-of-00086.safetensors b/model-00071-of-00086.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..4af64d149817c7fe68a6b217bd0c2a150fae38ca --- /dev/null +++ b/model-00071-of-00086.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:4e642d735217b70f5ff282497ef14493273aba6bfc497b8dae37f2703150edd3 +size 4631063720 diff --git a/model-00072-of-00086.safetensors b/model-00072-of-00086.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..2babcde8b0ea9c2a30066c11ae880ab360e2fe95 --- /dev/null +++ b/model-00072-of-00086.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:4e81614847972bb564b4b97a92de9938d6e5804142e9ef2909311e56fd90cfdd +size 4933097736 diff --git a/model-00073-of-00086.safetensors b/model-00073-of-00086.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..0595823544b23338a9d566011322933ea227f73f --- /dev/null +++ b/model-00073-of-00086.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:71527054bc6456b86c3e7a84e0787e55cfb7d6c34151c5731d82c48b5647fa7c +size 4631063720 diff --git a/model-00074-of-00086.safetensors b/model-00074-of-00086.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..963746f33762e49ebd2c1a5e3a8a6bc6ce6194b2 --- /dev/null +++ b/model-00074-of-00086.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:4094a00c597f9e403e53a3b1e443e3e05fff02551d2fe26f03137bf36ea2baee +size 4933097736 diff --git a/model-00075-of-00086.safetensors b/model-00075-of-00086.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..8590201c2bd155cc7268b513588d536005984c97 --- /dev/null +++ b/model-00075-of-00086.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:66f0f450b527de7804b285f54ead0adddb427a2a0064938a78799f0505e30309 +size 4631063720 diff --git a/model-00076-of-00086.safetensors b/model-00076-of-00086.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..c4d65f96adb5f605e775c8ac657d6a1282168813 --- /dev/null +++ b/model-00076-of-00086.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:30a8a69f040d6890c5c5a39df2001c6bd84cd25aa9c2466dd76653ab6dec1ae2 +size 4933097736 diff --git a/model-00077-of-00086.safetensors b/model-00077-of-00086.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..eaf34685b2e350dd425fd6e3b86fc47732b09441 --- /dev/null +++ b/model-00077-of-00086.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:203888977efdb8953dd1c51a10f6534c258866770a25b2e73e4fcc10445668ff +size 4631063720 diff --git a/model-00078-of-00086.safetensors b/model-00078-of-00086.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..c59868ab4763e896c29b5205667e7d1e8dae20b4 --- /dev/null +++ b/model-00078-of-00086.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:421a441c3c8af90bdf9373c9ad8bdb6ff5f46ab3b2832abef3aa8794e8a9b0fa +size 4933097736 diff --git a/model-00079-of-00086.safetensors b/model-00079-of-00086.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..f384d2f704fb2ae1afdb514e1cc879aefcee84b9 --- /dev/null +++ b/model-00079-of-00086.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7bb69eed700a2966484ab4a1a546480c95cd685607b873b643ef9c3b5f5de219 +size 4631063720 diff --git a/model-00080-of-00086.safetensors b/model-00080-of-00086.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..85fa18c4b2e7d5f3977db67a49a348044adc0ad1 --- /dev/null +++ b/model-00080-of-00086.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:187323e4debc4692e8397b0025236e2d89f7ec82e04345685d2e19e6e4a2cb24 +size 4933097736 diff --git a/model-00081-of-00086.safetensors b/model-00081-of-00086.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..5b4a88e2d53237c9f73b09ab70fe14d25fb644f9 --- /dev/null +++ b/model-00081-of-00086.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:eba015849eefd529f66a4bf4bc23aaa2d6735e0d71d596d6e569294f2eeec279 +size 4631063720 diff --git a/model-00082-of-00086.safetensors b/model-00082-of-00086.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..e49c4d583ef683a3ca893b8146fcaeb48854a963 --- /dev/null +++ b/model-00082-of-00086.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:4b14ecd39756e961f616dd10253002c8384abdd8c936689a099ef24e957ff470 +size 4933097736 diff --git a/model-00083-of-00086.safetensors b/model-00083-of-00086.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..0bad8be601f9acda5e75c72569643611a0a48077 --- /dev/null +++ b/model-00083-of-00086.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7cf56ce720476ec65aa3e86672b4e5973057a0a8be519f9f3f1b2a8e9fae31f8 +size 4631063720 diff --git a/model-00084-of-00086.safetensors b/model-00084-of-00086.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..f20efddb3e9d4b5328ba4c442c52d05b17c9f56d --- /dev/null +++ b/model-00084-of-00086.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a6c98fbe05470de4a4a861026ed0ab9b66369f6ee36f00dd4254644fffc04294 +size 4933097736 diff --git a/model-00085-of-00086.safetensors b/model-00085-of-00086.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..75f2f298d811d6379e957efd86b6b5d348b5ed2b --- /dev/null +++ b/model-00085-of-00086.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:02fa1076fef4a2cf4df1da18d4c4ff9bcfd2733e6f16c9c34d7b09ae5528f35b +size 4060600664 diff --git a/model-00086-of-00086.safetensors b/model-00086-of-00086.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..66056b0210948e827b80eb5fdce25bc9a9f39254 --- /dev/null +++ b/model-00086-of-00086.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:1b59a8da2692c90aa3024c77c6e1d5284b801b1aa8fd2a5a5546fb97d3aaaa34 +size 4202692736 diff --git a/model.safetensors.index.json b/model.safetensors.index.json new file mode 100644 index 0000000000000000000000000000000000000000..861a38388765478cc85524fef32cf0999e39c778 --- /dev/null +++ b/model.safetensors.index.json @@ -0,0 +1,2026 @@ +{ + "metadata": { + "total_size": 410099965952 + }, + "weight_map": { + "lm_head.weight": "model-00086-of-00086.safetensors", + "model.embed_tokens.weight": "model-00001-of-00086.safetensors", + "model.layers.0.input_layernorm.weight": "model-00002-of-00086.safetensors", + "model.layers.0.mlp.down_proj.weight": "model-00002-of-00086.safetensors", + "model.layers.0.mlp.down_proj.weight_scale": "model-00002-of-00086.safetensors", + "model.layers.0.mlp.gate_proj.weight": "model-00002-of-00086.safetensors", + "model.layers.0.mlp.gate_proj.weight_scale": "model-00002-of-00086.safetensors", + "model.layers.0.mlp.up_proj.weight": "model-00002-of-00086.safetensors", + "model.layers.0.mlp.up_proj.weight_scale": "model-00002-of-00086.safetensors", + "model.layers.0.post_attention_layernorm.weight": "model-00002-of-00086.safetensors", + "model.layers.0.self_attn.k_proj.weight": "model-00001-of-00086.safetensors", + "model.layers.0.self_attn.k_proj.weight_scale": "model-00001-of-00086.safetensors", + "model.layers.0.self_attn.o_proj.weight": "model-00001-of-00086.safetensors", + "model.layers.0.self_attn.o_proj.weight_scale": "model-00001-of-00086.safetensors", + "model.layers.0.self_attn.q_proj.weight": "model-00001-of-00086.safetensors", + "model.layers.0.self_attn.q_proj.weight_scale": "model-00001-of-00086.safetensors", + "model.layers.0.self_attn.v_proj.weight": "model-00001-of-00086.safetensors", + "model.layers.0.self_attn.v_proj.weight_scale": "model-00001-of-00086.safetensors", + "model.layers.1.input_layernorm.weight": "model-00003-of-00086.safetensors", + "model.layers.1.mlp.down_proj.weight": "model-00003-of-00086.safetensors", + "model.layers.1.mlp.down_proj.weight_scale": "model-00003-of-00086.safetensors", + "model.layers.1.mlp.gate_proj.weight": "model-00002-of-00086.safetensors", + "model.layers.1.mlp.gate_proj.weight_scale": "model-00002-of-00086.safetensors", + "model.layers.1.mlp.up_proj.weight": "model-00002-of-00086.safetensors", + "model.layers.1.mlp.up_proj.weight_scale": "model-00002-of-00086.safetensors", + "model.layers.1.post_attention_layernorm.weight": "model-00003-of-00086.safetensors", + "model.layers.1.self_attn.k_proj.weight": "model-00002-of-00086.safetensors", + "model.layers.1.self_attn.k_proj.weight_scale": "model-00002-of-00086.safetensors", + "model.layers.1.self_attn.o_proj.weight": "model-00002-of-00086.safetensors", + "model.layers.1.self_attn.o_proj.weight_scale": "model-00002-of-00086.safetensors", + "model.layers.1.self_attn.q_proj.weight": "model-00002-of-00086.safetensors", + "model.layers.1.self_attn.q_proj.weight_scale": "model-00002-of-00086.safetensors", + "model.layers.1.self_attn.v_proj.weight": "model-00002-of-00086.safetensors", + "model.layers.1.self_attn.v_proj.weight_scale": "model-00002-of-00086.safetensors", + "model.layers.10.input_layernorm.weight": "model-00009-of-00086.safetensors", + "model.layers.10.mlp.down_proj.weight": "model-00009-of-00086.safetensors", + "model.layers.10.mlp.down_proj.weight_scale": "model-00009-of-00086.safetensors", + "model.layers.10.mlp.gate_proj.weight": "model-00008-of-00086.safetensors", + "model.layers.10.mlp.gate_proj.weight_scale": "model-00008-of-00086.safetensors", + "model.layers.10.mlp.up_proj.weight": "model-00008-of-00086.safetensors", + "model.layers.10.mlp.up_proj.weight_scale": "model-00008-of-00086.safetensors", + "model.layers.10.post_attention_layernorm.weight": "model-00009-of-00086.safetensors", + "model.layers.10.self_attn.k_proj.weight": "model-00008-of-00086.safetensors", + "model.layers.10.self_attn.k_proj.weight_scale": "model-00008-of-00086.safetensors", + "model.layers.10.self_attn.o_proj.weight": "model-00008-of-00086.safetensors", + "model.layers.10.self_attn.o_proj.weight_scale": "model-00008-of-00086.safetensors", + "model.layers.10.self_attn.q_proj.weight": "model-00008-of-00086.safetensors", + "model.layers.10.self_attn.q_proj.weight_scale": "model-00008-of-00086.safetensors", + "model.layers.10.self_attn.v_proj.weight": "model-00008-of-00086.safetensors", + "model.layers.10.self_attn.v_proj.weight_scale": "model-00008-of-00086.safetensors", + "model.layers.100.input_layernorm.weight": "model-00069-of-00086.safetensors", + "model.layers.100.mlp.down_proj.weight": "model-00069-of-00086.safetensors", + "model.layers.100.mlp.down_proj.weight_scale": "model-00069-of-00086.safetensors", + "model.layers.100.mlp.gate_proj.weight": "model-00068-of-00086.safetensors", + "model.layers.100.mlp.gate_proj.weight_scale": "model-00068-of-00086.safetensors", + "model.layers.100.mlp.up_proj.weight": "model-00068-of-00086.safetensors", + "model.layers.100.mlp.up_proj.weight_scale": "model-00068-of-00086.safetensors", + "model.layers.100.post_attention_layernorm.weight": "model-00069-of-00086.safetensors", + "model.layers.100.self_attn.k_proj.weight": "model-00068-of-00086.safetensors", + "model.layers.100.self_attn.k_proj.weight_scale": "model-00068-of-00086.safetensors", + "model.layers.100.self_attn.o_proj.weight": "model-00068-of-00086.safetensors", + "model.layers.100.self_attn.o_proj.weight_scale": "model-00068-of-00086.safetensors", + "model.layers.100.self_attn.q_proj.weight": "model-00068-of-00086.safetensors", + "model.layers.100.self_attn.q_proj.weight_scale": "model-00068-of-00086.safetensors", + "model.layers.100.self_attn.v_proj.weight": "model-00068-of-00086.safetensors", + "model.layers.100.self_attn.v_proj.weight_scale": "model-00068-of-00086.safetensors", + "model.layers.101.input_layernorm.weight": "model-00069-of-00086.safetensors", + "model.layers.101.mlp.down_proj.weight": "model-00069-of-00086.safetensors", + "model.layers.101.mlp.down_proj.weight_scale": "model-00069-of-00086.safetensors", + "model.layers.101.mlp.gate_proj.weight": "model-00069-of-00086.safetensors", + "model.layers.101.mlp.gate_proj.weight_scale": "model-00069-of-00086.safetensors", + "model.layers.101.mlp.up_proj.weight": "model-00069-of-00086.safetensors", + "model.layers.101.mlp.up_proj.weight_scale": "model-00069-of-00086.safetensors", + "model.layers.101.post_attention_layernorm.weight": "model-00069-of-00086.safetensors", + "model.layers.101.self_attn.k_proj.weight": "model-00069-of-00086.safetensors", + "model.layers.101.self_attn.k_proj.weight_scale": "model-00069-of-00086.safetensors", + "model.layers.101.self_attn.o_proj.weight": "model-00069-of-00086.safetensors", + "model.layers.101.self_attn.o_proj.weight_scale": "model-00069-of-00086.safetensors", + "model.layers.101.self_attn.q_proj.weight": "model-00069-of-00086.safetensors", + "model.layers.101.self_attn.q_proj.weight_scale": "model-00069-of-00086.safetensors", + "model.layers.101.self_attn.v_proj.weight": "model-00069-of-00086.safetensors", + "model.layers.101.self_attn.v_proj.weight_scale": "model-00069-of-00086.safetensors", + "model.layers.102.input_layernorm.weight": "model-00070-of-00086.safetensors", + "model.layers.102.mlp.down_proj.weight": "model-00070-of-00086.safetensors", + "model.layers.102.mlp.down_proj.weight_scale": "model-00070-of-00086.safetensors", + "model.layers.102.mlp.gate_proj.weight": "model-00070-of-00086.safetensors", + "model.layers.102.mlp.gate_proj.weight_scale": "model-00070-of-00086.safetensors", + "model.layers.102.mlp.up_proj.weight": "model-00070-of-00086.safetensors", + "model.layers.102.mlp.up_proj.weight_scale": "model-00070-of-00086.safetensors", + "model.layers.102.post_attention_layernorm.weight": "model-00070-of-00086.safetensors", + "model.layers.102.self_attn.k_proj.weight": "model-00069-of-00086.safetensors", + "model.layers.102.self_attn.k_proj.weight_scale": "model-00069-of-00086.safetensors", + "model.layers.102.self_attn.o_proj.weight": "model-00069-of-00086.safetensors", + "model.layers.102.self_attn.o_proj.weight_scale": "model-00069-of-00086.safetensors", + "model.layers.102.self_attn.q_proj.weight": "model-00069-of-00086.safetensors", + "model.layers.102.self_attn.q_proj.weight_scale": "model-00069-of-00086.safetensors", + "model.layers.102.self_attn.v_proj.weight": "model-00069-of-00086.safetensors", + "model.layers.102.self_attn.v_proj.weight_scale": "model-00069-of-00086.safetensors", + "model.layers.103.input_layernorm.weight": "model-00071-of-00086.safetensors", + "model.layers.103.mlp.down_proj.weight": "model-00071-of-00086.safetensors", + "model.layers.103.mlp.down_proj.weight_scale": "model-00071-of-00086.safetensors", + "model.layers.103.mlp.gate_proj.weight": "model-00070-of-00086.safetensors", + "model.layers.103.mlp.gate_proj.weight_scale": "model-00070-of-00086.safetensors", + "model.layers.103.mlp.up_proj.weight": "model-00070-of-00086.safetensors", + "model.layers.103.mlp.up_proj.weight_scale": "model-00070-of-00086.safetensors", + "model.layers.103.post_attention_layernorm.weight": "model-00071-of-00086.safetensors", + "model.layers.103.self_attn.k_proj.weight": "model-00070-of-00086.safetensors", + "model.layers.103.self_attn.k_proj.weight_scale": "model-00070-of-00086.safetensors", + "model.layers.103.self_attn.o_proj.weight": "model-00070-of-00086.safetensors", + "model.layers.103.self_attn.o_proj.weight_scale": "model-00070-of-00086.safetensors", + "model.layers.103.self_attn.q_proj.weight": "model-00070-of-00086.safetensors", + "model.layers.103.self_attn.q_proj.weight_scale": "model-00070-of-00086.safetensors", + "model.layers.103.self_attn.v_proj.weight": "model-00070-of-00086.safetensors", + "model.layers.103.self_attn.v_proj.weight_scale": "model-00070-of-00086.safetensors", + "model.layers.104.input_layernorm.weight": "model-00071-of-00086.safetensors", + "model.layers.104.mlp.down_proj.weight": "model-00071-of-00086.safetensors", + "model.layers.104.mlp.down_proj.weight_scale": "model-00071-of-00086.safetensors", + "model.layers.104.mlp.gate_proj.weight": "model-00071-of-00086.safetensors", + "model.layers.104.mlp.gate_proj.weight_scale": "model-00071-of-00086.safetensors", + "model.layers.104.mlp.up_proj.weight": "model-00071-of-00086.safetensors", + "model.layers.104.mlp.up_proj.weight_scale": "model-00071-of-00086.safetensors", + "model.layers.104.post_attention_layernorm.weight": "model-00071-of-00086.safetensors", + "model.layers.104.self_attn.k_proj.weight": "model-00071-of-00086.safetensors", + "model.layers.104.self_attn.k_proj.weight_scale": "model-00071-of-00086.safetensors", + "model.layers.104.self_attn.o_proj.weight": "model-00071-of-00086.safetensors", + "model.layers.104.self_attn.o_proj.weight_scale": "model-00071-of-00086.safetensors", + "model.layers.104.self_attn.q_proj.weight": "model-00071-of-00086.safetensors", + "model.layers.104.self_attn.q_proj.weight_scale": "model-00071-of-00086.safetensors", + "model.layers.104.self_attn.v_proj.weight": "model-00071-of-00086.safetensors", + "model.layers.104.self_attn.v_proj.weight_scale": "model-00071-of-00086.safetensors", + "model.layers.105.input_layernorm.weight": "model-00072-of-00086.safetensors", + "model.layers.105.mlp.down_proj.weight": "model-00072-of-00086.safetensors", + "model.layers.105.mlp.down_proj.weight_scale": "model-00072-of-00086.safetensors", + "model.layers.105.mlp.gate_proj.weight": "model-00072-of-00086.safetensors", + "model.layers.105.mlp.gate_proj.weight_scale": "model-00072-of-00086.safetensors", + "model.layers.105.mlp.up_proj.weight": "model-00072-of-00086.safetensors", + "model.layers.105.mlp.up_proj.weight_scale": "model-00072-of-00086.safetensors", + "model.layers.105.post_attention_layernorm.weight": "model-00072-of-00086.safetensors", + "model.layers.105.self_attn.k_proj.weight": "model-00071-of-00086.safetensors", + "model.layers.105.self_attn.k_proj.weight_scale": "model-00071-of-00086.safetensors", + "model.layers.105.self_attn.o_proj.weight": "model-00071-of-00086.safetensors", + "model.layers.105.self_attn.o_proj.weight_scale": "model-00071-of-00086.safetensors", + "model.layers.105.self_attn.q_proj.weight": "model-00071-of-00086.safetensors", + "model.layers.105.self_attn.q_proj.weight_scale": "model-00071-of-00086.safetensors", + "model.layers.105.self_attn.v_proj.weight": "model-00071-of-00086.safetensors", + "model.layers.105.self_attn.v_proj.weight_scale": "model-00071-of-00086.safetensors", + "model.layers.106.input_layernorm.weight": "model-00073-of-00086.safetensors", + "model.layers.106.mlp.down_proj.weight": "model-00073-of-00086.safetensors", + "model.layers.106.mlp.down_proj.weight_scale": "model-00073-of-00086.safetensors", + "model.layers.106.mlp.gate_proj.weight": "model-00072-of-00086.safetensors", + "model.layers.106.mlp.gate_proj.weight_scale": "model-00072-of-00086.safetensors", + "model.layers.106.mlp.up_proj.weight": "model-00072-of-00086.safetensors", + "model.layers.106.mlp.up_proj.weight_scale": "model-00072-of-00086.safetensors", + "model.layers.106.post_attention_layernorm.weight": "model-00073-of-00086.safetensors", + "model.layers.106.self_attn.k_proj.weight": "model-00072-of-00086.safetensors", + "model.layers.106.self_attn.k_proj.weight_scale": "model-00072-of-00086.safetensors", + "model.layers.106.self_attn.o_proj.weight": "model-00072-of-00086.safetensors", + "model.layers.106.self_attn.o_proj.weight_scale": "model-00072-of-00086.safetensors", + "model.layers.106.self_attn.q_proj.weight": "model-00072-of-00086.safetensors", + "model.layers.106.self_attn.q_proj.weight_scale": "model-00072-of-00086.safetensors", + "model.layers.106.self_attn.v_proj.weight": "model-00072-of-00086.safetensors", + "model.layers.106.self_attn.v_proj.weight_scale": "model-00072-of-00086.safetensors", + "model.layers.107.input_layernorm.weight": "model-00073-of-00086.safetensors", + "model.layers.107.mlp.down_proj.weight": "model-00073-of-00086.safetensors", + "model.layers.107.mlp.down_proj.weight_scale": "model-00073-of-00086.safetensors", + "model.layers.107.mlp.gate_proj.weight": "model-00073-of-00086.safetensors", + "model.layers.107.mlp.gate_proj.weight_scale": "model-00073-of-00086.safetensors", + "model.layers.107.mlp.up_proj.weight": "model-00073-of-00086.safetensors", + "model.layers.107.mlp.up_proj.weight_scale": "model-00073-of-00086.safetensors", + "model.layers.107.post_attention_layernorm.weight": "model-00073-of-00086.safetensors", + "model.layers.107.self_attn.k_proj.weight": "model-00073-of-00086.safetensors", + "model.layers.107.self_attn.k_proj.weight_scale": "model-00073-of-00086.safetensors", + "model.layers.107.self_attn.o_proj.weight": "model-00073-of-00086.safetensors", + "model.layers.107.self_attn.o_proj.weight_scale": "model-00073-of-00086.safetensors", + "model.layers.107.self_attn.q_proj.weight": "model-00073-of-00086.safetensors", + "model.layers.107.self_attn.q_proj.weight_scale": "model-00073-of-00086.safetensors", + "model.layers.107.self_attn.v_proj.weight": "model-00073-of-00086.safetensors", + "model.layers.107.self_attn.v_proj.weight_scale": "model-00073-of-00086.safetensors", + "model.layers.108.input_layernorm.weight": "model-00074-of-00086.safetensors", + "model.layers.108.mlp.down_proj.weight": "model-00074-of-00086.safetensors", + "model.layers.108.mlp.down_proj.weight_scale": "model-00074-of-00086.safetensors", + "model.layers.108.mlp.gate_proj.weight": "model-00074-of-00086.safetensors", + "model.layers.108.mlp.gate_proj.weight_scale": "model-00074-of-00086.safetensors", + "model.layers.108.mlp.up_proj.weight": "model-00074-of-00086.safetensors", + "model.layers.108.mlp.up_proj.weight_scale": "model-00074-of-00086.safetensors", + "model.layers.108.post_attention_layernorm.weight": "model-00074-of-00086.safetensors", + "model.layers.108.self_attn.k_proj.weight": "model-00073-of-00086.safetensors", + "model.layers.108.self_attn.k_proj.weight_scale": "model-00073-of-00086.safetensors", + "model.layers.108.self_attn.o_proj.weight": "model-00073-of-00086.safetensors", + "model.layers.108.self_attn.o_proj.weight_scale": "model-00073-of-00086.safetensors", + "model.layers.108.self_attn.q_proj.weight": "model-00073-of-00086.safetensors", + "model.layers.108.self_attn.q_proj.weight_scale": "model-00073-of-00086.safetensors", + "model.layers.108.self_attn.v_proj.weight": "model-00073-of-00086.safetensors", + "model.layers.108.self_attn.v_proj.weight_scale": "model-00073-of-00086.safetensors", + "model.layers.109.input_layernorm.weight": "model-00075-of-00086.safetensors", + "model.layers.109.mlp.down_proj.weight": "model-00075-of-00086.safetensors", + "model.layers.109.mlp.down_proj.weight_scale": "model-00075-of-00086.safetensors", + "model.layers.109.mlp.gate_proj.weight": "model-00074-of-00086.safetensors", + "model.layers.109.mlp.gate_proj.weight_scale": "model-00074-of-00086.safetensors", + "model.layers.109.mlp.up_proj.weight": "model-00074-of-00086.safetensors", + "model.layers.109.mlp.up_proj.weight_scale": "model-00074-of-00086.safetensors", + "model.layers.109.post_attention_layernorm.weight": "model-00075-of-00086.safetensors", + "model.layers.109.self_attn.k_proj.weight": "model-00074-of-00086.safetensors", + "model.layers.109.self_attn.k_proj.weight_scale": "model-00074-of-00086.safetensors", + "model.layers.109.self_attn.o_proj.weight": "model-00074-of-00086.safetensors", + "model.layers.109.self_attn.o_proj.weight_scale": "model-00074-of-00086.safetensors", + "model.layers.109.self_attn.q_proj.weight": "model-00074-of-00086.safetensors", + "model.layers.109.self_attn.q_proj.weight_scale": "model-00074-of-00086.safetensors", + "model.layers.109.self_attn.v_proj.weight": "model-00074-of-00086.safetensors", + "model.layers.109.self_attn.v_proj.weight_scale": "model-00074-of-00086.safetensors", + "model.layers.11.input_layernorm.weight": "model-00009-of-00086.safetensors", + "model.layers.11.mlp.down_proj.weight": "model-00009-of-00086.safetensors", + "model.layers.11.mlp.down_proj.weight_scale": "model-00009-of-00086.safetensors", + "model.layers.11.mlp.gate_proj.weight": "model-00009-of-00086.safetensors", + "model.layers.11.mlp.gate_proj.weight_scale": "model-00009-of-00086.safetensors", + "model.layers.11.mlp.up_proj.weight": "model-00009-of-00086.safetensors", + "model.layers.11.mlp.up_proj.weight_scale": "model-00009-of-00086.safetensors", + "model.layers.11.post_attention_layernorm.weight": "model-00009-of-00086.safetensors", + "model.layers.11.self_attn.k_proj.weight": "model-00009-of-00086.safetensors", + "model.layers.11.self_attn.k_proj.weight_scale": "model-00009-of-00086.safetensors", + "model.layers.11.self_attn.o_proj.weight": "model-00009-of-00086.safetensors", + "model.layers.11.self_attn.o_proj.weight_scale": "model-00009-of-00086.safetensors", + "model.layers.11.self_attn.q_proj.weight": "model-00009-of-00086.safetensors", + "model.layers.11.self_attn.q_proj.weight_scale": "model-00009-of-00086.safetensors", + "model.layers.11.self_attn.v_proj.weight": "model-00009-of-00086.safetensors", + "model.layers.11.self_attn.v_proj.weight_scale": "model-00009-of-00086.safetensors", + "model.layers.110.input_layernorm.weight": "model-00075-of-00086.safetensors", + "model.layers.110.mlp.down_proj.weight": "model-00075-of-00086.safetensors", + "model.layers.110.mlp.down_proj.weight_scale": "model-00075-of-00086.safetensors", + "model.layers.110.mlp.gate_proj.weight": "model-00075-of-00086.safetensors", + "model.layers.110.mlp.gate_proj.weight_scale": "model-00075-of-00086.safetensors", + "model.layers.110.mlp.up_proj.weight": "model-00075-of-00086.safetensors", + "model.layers.110.mlp.up_proj.weight_scale": "model-00075-of-00086.safetensors", + "model.layers.110.post_attention_layernorm.weight": "model-00075-of-00086.safetensors", + "model.layers.110.self_attn.k_proj.weight": "model-00075-of-00086.safetensors", + "model.layers.110.self_attn.k_proj.weight_scale": "model-00075-of-00086.safetensors", + "model.layers.110.self_attn.o_proj.weight": "model-00075-of-00086.safetensors", + "model.layers.110.self_attn.o_proj.weight_scale": "model-00075-of-00086.safetensors", + "model.layers.110.self_attn.q_proj.weight": "model-00075-of-00086.safetensors", + "model.layers.110.self_attn.q_proj.weight_scale": "model-00075-of-00086.safetensors", + "model.layers.110.self_attn.v_proj.weight": "model-00075-of-00086.safetensors", + "model.layers.110.self_attn.v_proj.weight_scale": "model-00075-of-00086.safetensors", + "model.layers.111.input_layernorm.weight": "model-00076-of-00086.safetensors", + "model.layers.111.mlp.down_proj.weight": "model-00076-of-00086.safetensors", + "model.layers.111.mlp.down_proj.weight_scale": "model-00076-of-00086.safetensors", + "model.layers.111.mlp.gate_proj.weight": "model-00076-of-00086.safetensors", + "model.layers.111.mlp.gate_proj.weight_scale": "model-00076-of-00086.safetensors", + "model.layers.111.mlp.up_proj.weight": "model-00076-of-00086.safetensors", + "model.layers.111.mlp.up_proj.weight_scale": "model-00076-of-00086.safetensors", + "model.layers.111.post_attention_layernorm.weight": "model-00076-of-00086.safetensors", + "model.layers.111.self_attn.k_proj.weight": "model-00075-of-00086.safetensors", + "model.layers.111.self_attn.k_proj.weight_scale": "model-00075-of-00086.safetensors", + "model.layers.111.self_attn.o_proj.weight": "model-00075-of-00086.safetensors", + "model.layers.111.self_attn.o_proj.weight_scale": "model-00075-of-00086.safetensors", + "model.layers.111.self_attn.q_proj.weight": "model-00075-of-00086.safetensors", + "model.layers.111.self_attn.q_proj.weight_scale": "model-00075-of-00086.safetensors", + "model.layers.111.self_attn.v_proj.weight": "model-00075-of-00086.safetensors", + "model.layers.111.self_attn.v_proj.weight_scale": "model-00075-of-00086.safetensors", + "model.layers.112.input_layernorm.weight": "model-00077-of-00086.safetensors", + "model.layers.112.mlp.down_proj.weight": "model-00077-of-00086.safetensors", + "model.layers.112.mlp.down_proj.weight_scale": "model-00077-of-00086.safetensors", + "model.layers.112.mlp.gate_proj.weight": "model-00076-of-00086.safetensors", + "model.layers.112.mlp.gate_proj.weight_scale": "model-00076-of-00086.safetensors", + "model.layers.112.mlp.up_proj.weight": "model-00076-of-00086.safetensors", + "model.layers.112.mlp.up_proj.weight_scale": "model-00076-of-00086.safetensors", + "model.layers.112.post_attention_layernorm.weight": "model-00077-of-00086.safetensors", + "model.layers.112.self_attn.k_proj.weight": "model-00076-of-00086.safetensors", + "model.layers.112.self_attn.k_proj.weight_scale": "model-00076-of-00086.safetensors", + "model.layers.112.self_attn.o_proj.weight": "model-00076-of-00086.safetensors", + "model.layers.112.self_attn.o_proj.weight_scale": "model-00076-of-00086.safetensors", + "model.layers.112.self_attn.q_proj.weight": "model-00076-of-00086.safetensors", + "model.layers.112.self_attn.q_proj.weight_scale": "model-00076-of-00086.safetensors", + "model.layers.112.self_attn.v_proj.weight": "model-00076-of-00086.safetensors", + "model.layers.112.self_attn.v_proj.weight_scale": "model-00076-of-00086.safetensors", + "model.layers.113.input_layernorm.weight": "model-00077-of-00086.safetensors", + "model.layers.113.mlp.down_proj.weight": "model-00077-of-00086.safetensors", + "model.layers.113.mlp.down_proj.weight_scale": "model-00077-of-00086.safetensors", + "model.layers.113.mlp.gate_proj.weight": "model-00077-of-00086.safetensors", + "model.layers.113.mlp.gate_proj.weight_scale": "model-00077-of-00086.safetensors", + "model.layers.113.mlp.up_proj.weight": "model-00077-of-00086.safetensors", + "model.layers.113.mlp.up_proj.weight_scale": "model-00077-of-00086.safetensors", + "model.layers.113.post_attention_layernorm.weight": "model-00077-of-00086.safetensors", + "model.layers.113.self_attn.k_proj.weight": "model-00077-of-00086.safetensors", + "model.layers.113.self_attn.k_proj.weight_scale": "model-00077-of-00086.safetensors", + "model.layers.113.self_attn.o_proj.weight": "model-00077-of-00086.safetensors", + "model.layers.113.self_attn.o_proj.weight_scale": "model-00077-of-00086.safetensors", + "model.layers.113.self_attn.q_proj.weight": "model-00077-of-00086.safetensors", + "model.layers.113.self_attn.q_proj.weight_scale": "model-00077-of-00086.safetensors", + "model.layers.113.self_attn.v_proj.weight": "model-00077-of-00086.safetensors", + "model.layers.113.self_attn.v_proj.weight_scale": "model-00077-of-00086.safetensors", + "model.layers.114.input_layernorm.weight": "model-00078-of-00086.safetensors", + "model.layers.114.mlp.down_proj.weight": "model-00078-of-00086.safetensors", + "model.layers.114.mlp.down_proj.weight_scale": "model-00078-of-00086.safetensors", + "model.layers.114.mlp.gate_proj.weight": "model-00078-of-00086.safetensors", + "model.layers.114.mlp.gate_proj.weight_scale": "model-00078-of-00086.safetensors", + "model.layers.114.mlp.up_proj.weight": "model-00078-of-00086.safetensors", + "model.layers.114.mlp.up_proj.weight_scale": "model-00078-of-00086.safetensors", + "model.layers.114.post_attention_layernorm.weight": "model-00078-of-00086.safetensors", + "model.layers.114.self_attn.k_proj.weight": "model-00077-of-00086.safetensors", + "model.layers.114.self_attn.k_proj.weight_scale": "model-00077-of-00086.safetensors", + "model.layers.114.self_attn.o_proj.weight": "model-00077-of-00086.safetensors", + "model.layers.114.self_attn.o_proj.weight_scale": "model-00077-of-00086.safetensors", + "model.layers.114.self_attn.q_proj.weight": "model-00077-of-00086.safetensors", + "model.layers.114.self_attn.q_proj.weight_scale": "model-00077-of-00086.safetensors", + "model.layers.114.self_attn.v_proj.weight": "model-00077-of-00086.safetensors", + "model.layers.114.self_attn.v_proj.weight_scale": "model-00077-of-00086.safetensors", + "model.layers.115.input_layernorm.weight": "model-00079-of-00086.safetensors", + "model.layers.115.mlp.down_proj.weight": "model-00079-of-00086.safetensors", + "model.layers.115.mlp.down_proj.weight_scale": "model-00079-of-00086.safetensors", + "model.layers.115.mlp.gate_proj.weight": "model-00078-of-00086.safetensors", + "model.layers.115.mlp.gate_proj.weight_scale": "model-00078-of-00086.safetensors", + "model.layers.115.mlp.up_proj.weight": "model-00078-of-00086.safetensors", + "model.layers.115.mlp.up_proj.weight_scale": "model-00078-of-00086.safetensors", + "model.layers.115.post_attention_layernorm.weight": "model-00079-of-00086.safetensors", + "model.layers.115.self_attn.k_proj.weight": "model-00078-of-00086.safetensors", + "model.layers.115.self_attn.k_proj.weight_scale": "model-00078-of-00086.safetensors", + "model.layers.115.self_attn.o_proj.weight": "model-00078-of-00086.safetensors", + "model.layers.115.self_attn.o_proj.weight_scale": "model-00078-of-00086.safetensors", + "model.layers.115.self_attn.q_proj.weight": "model-00078-of-00086.safetensors", + "model.layers.115.self_attn.q_proj.weight_scale": "model-00078-of-00086.safetensors", + "model.layers.115.self_attn.v_proj.weight": "model-00078-of-00086.safetensors", + "model.layers.115.self_attn.v_proj.weight_scale": "model-00078-of-00086.safetensors", + "model.layers.116.input_layernorm.weight": "model-00079-of-00086.safetensors", + "model.layers.116.mlp.down_proj.weight": "model-00079-of-00086.safetensors", + "model.layers.116.mlp.down_proj.weight_scale": "model-00079-of-00086.safetensors", + "model.layers.116.mlp.gate_proj.weight": "model-00079-of-00086.safetensors", + "model.layers.116.mlp.gate_proj.weight_scale": "model-00079-of-00086.safetensors", + "model.layers.116.mlp.up_proj.weight": "model-00079-of-00086.safetensors", + "model.layers.116.mlp.up_proj.weight_scale": "model-00079-of-00086.safetensors", + "model.layers.116.post_attention_layernorm.weight": "model-00079-of-00086.safetensors", + "model.layers.116.self_attn.k_proj.weight": "model-00079-of-00086.safetensors", + "model.layers.116.self_attn.k_proj.weight_scale": "model-00079-of-00086.safetensors", + "model.layers.116.self_attn.o_proj.weight": "model-00079-of-00086.safetensors", + "model.layers.116.self_attn.o_proj.weight_scale": "model-00079-of-00086.safetensors", + "model.layers.116.self_attn.q_proj.weight": "model-00079-of-00086.safetensors", + "model.layers.116.self_attn.q_proj.weight_scale": "model-00079-of-00086.safetensors", + "model.layers.116.self_attn.v_proj.weight": "model-00079-of-00086.safetensors", + "model.layers.116.self_attn.v_proj.weight_scale": "model-00079-of-00086.safetensors", + "model.layers.117.input_layernorm.weight": "model-00080-of-00086.safetensors", + "model.layers.117.mlp.down_proj.weight": "model-00080-of-00086.safetensors", + "model.layers.117.mlp.down_proj.weight_scale": "model-00080-of-00086.safetensors", + "model.layers.117.mlp.gate_proj.weight": "model-00080-of-00086.safetensors", + "model.layers.117.mlp.gate_proj.weight_scale": "model-00080-of-00086.safetensors", + "model.layers.117.mlp.up_proj.weight": "model-00080-of-00086.safetensors", + "model.layers.117.mlp.up_proj.weight_scale": "model-00080-of-00086.safetensors", + "model.layers.117.post_attention_layernorm.weight": "model-00080-of-00086.safetensors", + "model.layers.117.self_attn.k_proj.weight": "model-00079-of-00086.safetensors", + "model.layers.117.self_attn.k_proj.weight_scale": "model-00079-of-00086.safetensors", + "model.layers.117.self_attn.o_proj.weight": "model-00079-of-00086.safetensors", + "model.layers.117.self_attn.o_proj.weight_scale": "model-00079-of-00086.safetensors", + "model.layers.117.self_attn.q_proj.weight": "model-00079-of-00086.safetensors", + "model.layers.117.self_attn.q_proj.weight_scale": "model-00079-of-00086.safetensors", + "model.layers.117.self_attn.v_proj.weight": "model-00079-of-00086.safetensors", + "model.layers.117.self_attn.v_proj.weight_scale": "model-00079-of-00086.safetensors", + "model.layers.118.input_layernorm.weight": "model-00081-of-00086.safetensors", + "model.layers.118.mlp.down_proj.weight": "model-00081-of-00086.safetensors", + "model.layers.118.mlp.down_proj.weight_scale": "model-00081-of-00086.safetensors", + "model.layers.118.mlp.gate_proj.weight": "model-00080-of-00086.safetensors", + "model.layers.118.mlp.gate_proj.weight_scale": "model-00080-of-00086.safetensors", + "model.layers.118.mlp.up_proj.weight": "model-00080-of-00086.safetensors", + "model.layers.118.mlp.up_proj.weight_scale": "model-00080-of-00086.safetensors", + "model.layers.118.post_attention_layernorm.weight": "model-00081-of-00086.safetensors", + "model.layers.118.self_attn.k_proj.weight": "model-00080-of-00086.safetensors", + "model.layers.118.self_attn.k_proj.weight_scale": "model-00080-of-00086.safetensors", + "model.layers.118.self_attn.o_proj.weight": "model-00080-of-00086.safetensors", + "model.layers.118.self_attn.o_proj.weight_scale": "model-00080-of-00086.safetensors", + "model.layers.118.self_attn.q_proj.weight": "model-00080-of-00086.safetensors", + "model.layers.118.self_attn.q_proj.weight_scale": "model-00080-of-00086.safetensors", + "model.layers.118.self_attn.v_proj.weight": "model-00080-of-00086.safetensors", + "model.layers.118.self_attn.v_proj.weight_scale": "model-00080-of-00086.safetensors", + "model.layers.119.input_layernorm.weight": "model-00081-of-00086.safetensors", + "model.layers.119.mlp.down_proj.weight": "model-00081-of-00086.safetensors", + "model.layers.119.mlp.down_proj.weight_scale": "model-00081-of-00086.safetensors", + "model.layers.119.mlp.gate_proj.weight": "model-00081-of-00086.safetensors", + "model.layers.119.mlp.gate_proj.weight_scale": "model-00081-of-00086.safetensors", + "model.layers.119.mlp.up_proj.weight": "model-00081-of-00086.safetensors", + "model.layers.119.mlp.up_proj.weight_scale": "model-00081-of-00086.safetensors", + "model.layers.119.post_attention_layernorm.weight": "model-00081-of-00086.safetensors", + "model.layers.119.self_attn.k_proj.weight": "model-00081-of-00086.safetensors", + "model.layers.119.self_attn.k_proj.weight_scale": "model-00081-of-00086.safetensors", + "model.layers.119.self_attn.o_proj.weight": "model-00081-of-00086.safetensors", + "model.layers.119.self_attn.o_proj.weight_scale": "model-00081-of-00086.safetensors", + "model.layers.119.self_attn.q_proj.weight": "model-00081-of-00086.safetensors", + "model.layers.119.self_attn.q_proj.weight_scale": "model-00081-of-00086.safetensors", + "model.layers.119.self_attn.v_proj.weight": "model-00081-of-00086.safetensors", + "model.layers.119.self_attn.v_proj.weight_scale": "model-00081-of-00086.safetensors", + "model.layers.12.input_layernorm.weight": "model-00010-of-00086.safetensors", + "model.layers.12.mlp.down_proj.weight": "model-00010-of-00086.safetensors", + "model.layers.12.mlp.down_proj.weight_scale": "model-00010-of-00086.safetensors", + "model.layers.12.mlp.gate_proj.weight": "model-00010-of-00086.safetensors", + "model.layers.12.mlp.gate_proj.weight_scale": "model-00010-of-00086.safetensors", + "model.layers.12.mlp.up_proj.weight": "model-00010-of-00086.safetensors", + "model.layers.12.mlp.up_proj.weight_scale": "model-00010-of-00086.safetensors", + "model.layers.12.post_attention_layernorm.weight": "model-00010-of-00086.safetensors", + "model.layers.12.self_attn.k_proj.weight": "model-00009-of-00086.safetensors", + "model.layers.12.self_attn.k_proj.weight_scale": "model-00009-of-00086.safetensors", + "model.layers.12.self_attn.o_proj.weight": "model-00009-of-00086.safetensors", + "model.layers.12.self_attn.o_proj.weight_scale": "model-00009-of-00086.safetensors", + "model.layers.12.self_attn.q_proj.weight": "model-00009-of-00086.safetensors", + "model.layers.12.self_attn.q_proj.weight_scale": "model-00009-of-00086.safetensors", + "model.layers.12.self_attn.v_proj.weight": "model-00009-of-00086.safetensors", + "model.layers.12.self_attn.v_proj.weight_scale": "model-00009-of-00086.safetensors", + "model.layers.120.input_layernorm.weight": "model-00082-of-00086.safetensors", + "model.layers.120.mlp.down_proj.weight": "model-00082-of-00086.safetensors", + "model.layers.120.mlp.down_proj.weight_scale": "model-00082-of-00086.safetensors", + "model.layers.120.mlp.gate_proj.weight": "model-00082-of-00086.safetensors", + "model.layers.120.mlp.gate_proj.weight_scale": "model-00082-of-00086.safetensors", + "model.layers.120.mlp.up_proj.weight": "model-00082-of-00086.safetensors", + "model.layers.120.mlp.up_proj.weight_scale": "model-00082-of-00086.safetensors", + "model.layers.120.post_attention_layernorm.weight": "model-00082-of-00086.safetensors", + "model.layers.120.self_attn.k_proj.weight": "model-00081-of-00086.safetensors", + "model.layers.120.self_attn.k_proj.weight_scale": "model-00081-of-00086.safetensors", + "model.layers.120.self_attn.o_proj.weight": "model-00081-of-00086.safetensors", + "model.layers.120.self_attn.o_proj.weight_scale": "model-00081-of-00086.safetensors", + "model.layers.120.self_attn.q_proj.weight": "model-00081-of-00086.safetensors", + "model.layers.120.self_attn.q_proj.weight_scale": "model-00081-of-00086.safetensors", + "model.layers.120.self_attn.v_proj.weight": "model-00081-of-00086.safetensors", + "model.layers.120.self_attn.v_proj.weight_scale": "model-00081-of-00086.safetensors", + "model.layers.121.input_layernorm.weight": "model-00083-of-00086.safetensors", + "model.layers.121.mlp.down_proj.weight": "model-00083-of-00086.safetensors", + "model.layers.121.mlp.down_proj.weight_scale": "model-00083-of-00086.safetensors", + "model.layers.121.mlp.gate_proj.weight": "model-00082-of-00086.safetensors", + "model.layers.121.mlp.gate_proj.weight_scale": "model-00082-of-00086.safetensors", + "model.layers.121.mlp.up_proj.weight": "model-00082-of-00086.safetensors", + "model.layers.121.mlp.up_proj.weight_scale": "model-00082-of-00086.safetensors", + "model.layers.121.post_attention_layernorm.weight": "model-00083-of-00086.safetensors", + "model.layers.121.self_attn.k_proj.weight": "model-00082-of-00086.safetensors", + "model.layers.121.self_attn.k_proj.weight_scale": "model-00082-of-00086.safetensors", + "model.layers.121.self_attn.o_proj.weight": "model-00082-of-00086.safetensors", + "model.layers.121.self_attn.o_proj.weight_scale": "model-00082-of-00086.safetensors", + "model.layers.121.self_attn.q_proj.weight": "model-00082-of-00086.safetensors", + "model.layers.121.self_attn.q_proj.weight_scale": "model-00082-of-00086.safetensors", + "model.layers.121.self_attn.v_proj.weight": "model-00082-of-00086.safetensors", + "model.layers.121.self_attn.v_proj.weight_scale": "model-00082-of-00086.safetensors", + "model.layers.122.input_layernorm.weight": "model-00083-of-00086.safetensors", + "model.layers.122.mlp.down_proj.weight": "model-00083-of-00086.safetensors", + "model.layers.122.mlp.down_proj.weight_scale": "model-00083-of-00086.safetensors", + "model.layers.122.mlp.gate_proj.weight": "model-00083-of-00086.safetensors", + "model.layers.122.mlp.gate_proj.weight_scale": "model-00083-of-00086.safetensors", + "model.layers.122.mlp.up_proj.weight": "model-00083-of-00086.safetensors", + "model.layers.122.mlp.up_proj.weight_scale": "model-00083-of-00086.safetensors", + "model.layers.122.post_attention_layernorm.weight": "model-00083-of-00086.safetensors", + "model.layers.122.self_attn.k_proj.weight": "model-00083-of-00086.safetensors", + "model.layers.122.self_attn.k_proj.weight_scale": "model-00083-of-00086.safetensors", + "model.layers.122.self_attn.o_proj.weight": "model-00083-of-00086.safetensors", + "model.layers.122.self_attn.o_proj.weight_scale": "model-00083-of-00086.safetensors", + "model.layers.122.self_attn.q_proj.weight": "model-00083-of-00086.safetensors", + "model.layers.122.self_attn.q_proj.weight_scale": "model-00083-of-00086.safetensors", + "model.layers.122.self_attn.v_proj.weight": "model-00083-of-00086.safetensors", + "model.layers.122.self_attn.v_proj.weight_scale": "model-00083-of-00086.safetensors", + "model.layers.123.input_layernorm.weight": "model-00084-of-00086.safetensors", + "model.layers.123.mlp.down_proj.weight": "model-00084-of-00086.safetensors", + "model.layers.123.mlp.down_proj.weight_scale": "model-00084-of-00086.safetensors", + "model.layers.123.mlp.gate_proj.weight": "model-00084-of-00086.safetensors", + "model.layers.123.mlp.gate_proj.weight_scale": "model-00084-of-00086.safetensors", + "model.layers.123.mlp.up_proj.weight": "model-00084-of-00086.safetensors", + "model.layers.123.mlp.up_proj.weight_scale": "model-00084-of-00086.safetensors", + "model.layers.123.post_attention_layernorm.weight": "model-00084-of-00086.safetensors", + "model.layers.123.self_attn.k_proj.weight": "model-00083-of-00086.safetensors", + "model.layers.123.self_attn.k_proj.weight_scale": "model-00083-of-00086.safetensors", + "model.layers.123.self_attn.o_proj.weight": "model-00083-of-00086.safetensors", + "model.layers.123.self_attn.o_proj.weight_scale": "model-00083-of-00086.safetensors", + "model.layers.123.self_attn.q_proj.weight": "model-00083-of-00086.safetensors", + "model.layers.123.self_attn.q_proj.weight_scale": "model-00083-of-00086.safetensors", + "model.layers.123.self_attn.v_proj.weight": "model-00083-of-00086.safetensors", + "model.layers.123.self_attn.v_proj.weight_scale": "model-00083-of-00086.safetensors", + "model.layers.124.input_layernorm.weight": "model-00085-of-00086.safetensors", + "model.layers.124.mlp.down_proj.weight": "model-00085-of-00086.safetensors", + "model.layers.124.mlp.down_proj.weight_scale": "model-00085-of-00086.safetensors", + "model.layers.124.mlp.gate_proj.weight": "model-00084-of-00086.safetensors", + "model.layers.124.mlp.gate_proj.weight_scale": "model-00084-of-00086.safetensors", + "model.layers.124.mlp.up_proj.weight": "model-00084-of-00086.safetensors", + "model.layers.124.mlp.up_proj.weight_scale": "model-00084-of-00086.safetensors", + "model.layers.124.post_attention_layernorm.weight": "model-00085-of-00086.safetensors", + "model.layers.124.self_attn.k_proj.weight": "model-00084-of-00086.safetensors", + "model.layers.124.self_attn.k_proj.weight_scale": "model-00084-of-00086.safetensors", + "model.layers.124.self_attn.o_proj.weight": "model-00084-of-00086.safetensors", + "model.layers.124.self_attn.o_proj.weight_scale": "model-00084-of-00086.safetensors", + "model.layers.124.self_attn.q_proj.weight": "model-00084-of-00086.safetensors", + "model.layers.124.self_attn.q_proj.weight_scale": "model-00084-of-00086.safetensors", + "model.layers.124.self_attn.v_proj.weight": "model-00084-of-00086.safetensors", + "model.layers.124.self_attn.v_proj.weight_scale": "model-00084-of-00086.safetensors", + "model.layers.125.input_layernorm.weight": "model-00085-of-00086.safetensors", + "model.layers.125.mlp.down_proj.weight": "model-00085-of-00086.safetensors", + "model.layers.125.mlp.down_proj.weight_scale": "model-00085-of-00086.safetensors", + "model.layers.125.mlp.gate_proj.weight": "model-00085-of-00086.safetensors", + "model.layers.125.mlp.gate_proj.weight_scale": "model-00085-of-00086.safetensors", + "model.layers.125.mlp.up_proj.weight": "model-00085-of-00086.safetensors", + "model.layers.125.mlp.up_proj.weight_scale": "model-00085-of-00086.safetensors", + "model.layers.125.post_attention_layernorm.weight": "model-00085-of-00086.safetensors", + "model.layers.125.self_attn.k_proj.weight": "model-00085-of-00086.safetensors", + "model.layers.125.self_attn.k_proj.weight_scale": "model-00085-of-00086.safetensors", + "model.layers.125.self_attn.o_proj.weight": "model-00085-of-00086.safetensors", + "model.layers.125.self_attn.o_proj.weight_scale": "model-00085-of-00086.safetensors", + "model.layers.125.self_attn.q_proj.weight": "model-00085-of-00086.safetensors", + "model.layers.125.self_attn.q_proj.weight_scale": "model-00085-of-00086.safetensors", + "model.layers.125.self_attn.v_proj.weight": "model-00085-of-00086.safetensors", + "model.layers.125.self_attn.v_proj.weight_scale": "model-00085-of-00086.safetensors", + "model.layers.13.input_layernorm.weight": "model-00011-of-00086.safetensors", + "model.layers.13.mlp.down_proj.weight": "model-00011-of-00086.safetensors", + "model.layers.13.mlp.down_proj.weight_scale": "model-00011-of-00086.safetensors", + "model.layers.13.mlp.gate_proj.weight": "model-00010-of-00086.safetensors", + "model.layers.13.mlp.gate_proj.weight_scale": "model-00010-of-00086.safetensors", + "model.layers.13.mlp.up_proj.weight": "model-00010-of-00086.safetensors", + "model.layers.13.mlp.up_proj.weight_scale": "model-00010-of-00086.safetensors", + "model.layers.13.post_attention_layernorm.weight": "model-00011-of-00086.safetensors", + "model.layers.13.self_attn.k_proj.weight": "model-00010-of-00086.safetensors", + "model.layers.13.self_attn.k_proj.weight_scale": "model-00010-of-00086.safetensors", + "model.layers.13.self_attn.o_proj.weight": "model-00010-of-00086.safetensors", + "model.layers.13.self_attn.o_proj.weight_scale": "model-00010-of-00086.safetensors", + "model.layers.13.self_attn.q_proj.weight": "model-00010-of-00086.safetensors", + "model.layers.13.self_attn.q_proj.weight_scale": "model-00010-of-00086.safetensors", + "model.layers.13.self_attn.v_proj.weight": "model-00010-of-00086.safetensors", + "model.layers.13.self_attn.v_proj.weight_scale": "model-00010-of-00086.safetensors", + "model.layers.14.input_layernorm.weight": "model-00011-of-00086.safetensors", + "model.layers.14.mlp.down_proj.weight": "model-00011-of-00086.safetensors", + "model.layers.14.mlp.down_proj.weight_scale": "model-00011-of-00086.safetensors", + "model.layers.14.mlp.gate_proj.weight": "model-00011-of-00086.safetensors", + "model.layers.14.mlp.gate_proj.weight_scale": "model-00011-of-00086.safetensors", + "model.layers.14.mlp.up_proj.weight": "model-00011-of-00086.safetensors", + "model.layers.14.mlp.up_proj.weight_scale": "model-00011-of-00086.safetensors", + "model.layers.14.post_attention_layernorm.weight": "model-00011-of-00086.safetensors", + "model.layers.14.self_attn.k_proj.weight": "model-00011-of-00086.safetensors", + "model.layers.14.self_attn.k_proj.weight_scale": "model-00011-of-00086.safetensors", + "model.layers.14.self_attn.o_proj.weight": "model-00011-of-00086.safetensors", + "model.layers.14.self_attn.o_proj.weight_scale": "model-00011-of-00086.safetensors", + "model.layers.14.self_attn.q_proj.weight": "model-00011-of-00086.safetensors", + "model.layers.14.self_attn.q_proj.weight_scale": "model-00011-of-00086.safetensors", + "model.layers.14.self_attn.v_proj.weight": "model-00011-of-00086.safetensors", + "model.layers.14.self_attn.v_proj.weight_scale": "model-00011-of-00086.safetensors", + "model.layers.15.input_layernorm.weight": "model-00012-of-00086.safetensors", + "model.layers.15.mlp.down_proj.weight": "model-00012-of-00086.safetensors", + "model.layers.15.mlp.down_proj.weight_scale": "model-00012-of-00086.safetensors", + "model.layers.15.mlp.gate_proj.weight": "model-00012-of-00086.safetensors", + "model.layers.15.mlp.gate_proj.weight_scale": "model-00012-of-00086.safetensors", + "model.layers.15.mlp.up_proj.weight": "model-00012-of-00086.safetensors", + "model.layers.15.mlp.up_proj.weight_scale": "model-00012-of-00086.safetensors", + "model.layers.15.post_attention_layernorm.weight": "model-00012-of-00086.safetensors", + "model.layers.15.self_attn.k_proj.weight": "model-00011-of-00086.safetensors", + "model.layers.15.self_attn.k_proj.weight_scale": "model-00011-of-00086.safetensors", + "model.layers.15.self_attn.o_proj.weight": "model-00011-of-00086.safetensors", + "model.layers.15.self_attn.o_proj.weight_scale": "model-00011-of-00086.safetensors", + "model.layers.15.self_attn.q_proj.weight": "model-00011-of-00086.safetensors", + "model.layers.15.self_attn.q_proj.weight_scale": "model-00011-of-00086.safetensors", + "model.layers.15.self_attn.v_proj.weight": "model-00011-of-00086.safetensors", + "model.layers.15.self_attn.v_proj.weight_scale": "model-00011-of-00086.safetensors", + "model.layers.16.input_layernorm.weight": "model-00013-of-00086.safetensors", + "model.layers.16.mlp.down_proj.weight": "model-00013-of-00086.safetensors", + "model.layers.16.mlp.down_proj.weight_scale": "model-00013-of-00086.safetensors", + "model.layers.16.mlp.gate_proj.weight": "model-00012-of-00086.safetensors", + "model.layers.16.mlp.gate_proj.weight_scale": "model-00012-of-00086.safetensors", + "model.layers.16.mlp.up_proj.weight": "model-00012-of-00086.safetensors", + "model.layers.16.mlp.up_proj.weight_scale": "model-00012-of-00086.safetensors", + "model.layers.16.post_attention_layernorm.weight": "model-00013-of-00086.safetensors", + "model.layers.16.self_attn.k_proj.weight": "model-00012-of-00086.safetensors", + "model.layers.16.self_attn.k_proj.weight_scale": "model-00012-of-00086.safetensors", + "model.layers.16.self_attn.o_proj.weight": "model-00012-of-00086.safetensors", + "model.layers.16.self_attn.o_proj.weight_scale": "model-00012-of-00086.safetensors", + "model.layers.16.self_attn.q_proj.weight": "model-00012-of-00086.safetensors", + "model.layers.16.self_attn.q_proj.weight_scale": "model-00012-of-00086.safetensors", + "model.layers.16.self_attn.v_proj.weight": "model-00012-of-00086.safetensors", + "model.layers.16.self_attn.v_proj.weight_scale": "model-00012-of-00086.safetensors", + "model.layers.17.input_layernorm.weight": "model-00013-of-00086.safetensors", + "model.layers.17.mlp.down_proj.weight": "model-00013-of-00086.safetensors", + "model.layers.17.mlp.down_proj.weight_scale": "model-00013-of-00086.safetensors", + "model.layers.17.mlp.gate_proj.weight": "model-00013-of-00086.safetensors", + "model.layers.17.mlp.gate_proj.weight_scale": "model-00013-of-00086.safetensors", + "model.layers.17.mlp.up_proj.weight": "model-00013-of-00086.safetensors", + "model.layers.17.mlp.up_proj.weight_scale": "model-00013-of-00086.safetensors", + "model.layers.17.post_attention_layernorm.weight": "model-00013-of-00086.safetensors", + "model.layers.17.self_attn.k_proj.weight": "model-00013-of-00086.safetensors", + "model.layers.17.self_attn.k_proj.weight_scale": "model-00013-of-00086.safetensors", + "model.layers.17.self_attn.o_proj.weight": "model-00013-of-00086.safetensors", + "model.layers.17.self_attn.o_proj.weight_scale": "model-00013-of-00086.safetensors", + "model.layers.17.self_attn.q_proj.weight": "model-00013-of-00086.safetensors", + "model.layers.17.self_attn.q_proj.weight_scale": "model-00013-of-00086.safetensors", + "model.layers.17.self_attn.v_proj.weight": "model-00013-of-00086.safetensors", + "model.layers.17.self_attn.v_proj.weight_scale": "model-00013-of-00086.safetensors", + "model.layers.18.input_layernorm.weight": "model-00014-of-00086.safetensors", + "model.layers.18.mlp.down_proj.weight": "model-00014-of-00086.safetensors", + "model.layers.18.mlp.down_proj.weight_scale": "model-00014-of-00086.safetensors", + "model.layers.18.mlp.gate_proj.weight": "model-00014-of-00086.safetensors", + "model.layers.18.mlp.gate_proj.weight_scale": "model-00014-of-00086.safetensors", + "model.layers.18.mlp.up_proj.weight": "model-00014-of-00086.safetensors", + "model.layers.18.mlp.up_proj.weight_scale": "model-00014-of-00086.safetensors", + "model.layers.18.post_attention_layernorm.weight": "model-00014-of-00086.safetensors", + "model.layers.18.self_attn.k_proj.weight": "model-00013-of-00086.safetensors", + "model.layers.18.self_attn.k_proj.weight_scale": "model-00013-of-00086.safetensors", + "model.layers.18.self_attn.o_proj.weight": "model-00013-of-00086.safetensors", + "model.layers.18.self_attn.o_proj.weight_scale": "model-00013-of-00086.safetensors", + "model.layers.18.self_attn.q_proj.weight": "model-00013-of-00086.safetensors", + "model.layers.18.self_attn.q_proj.weight_scale": "model-00013-of-00086.safetensors", + "model.layers.18.self_attn.v_proj.weight": "model-00013-of-00086.safetensors", + "model.layers.18.self_attn.v_proj.weight_scale": "model-00013-of-00086.safetensors", + "model.layers.19.input_layernorm.weight": "model-00015-of-00086.safetensors", + "model.layers.19.mlp.down_proj.weight": "model-00015-of-00086.safetensors", + "model.layers.19.mlp.down_proj.weight_scale": "model-00015-of-00086.safetensors", + "model.layers.19.mlp.gate_proj.weight": "model-00014-of-00086.safetensors", + "model.layers.19.mlp.gate_proj.weight_scale": "model-00014-of-00086.safetensors", + "model.layers.19.mlp.up_proj.weight": "model-00014-of-00086.safetensors", + "model.layers.19.mlp.up_proj.weight_scale": "model-00014-of-00086.safetensors", + "model.layers.19.post_attention_layernorm.weight": "model-00015-of-00086.safetensors", + "model.layers.19.self_attn.k_proj.weight": "model-00014-of-00086.safetensors", + "model.layers.19.self_attn.k_proj.weight_scale": "model-00014-of-00086.safetensors", + "model.layers.19.self_attn.o_proj.weight": "model-00014-of-00086.safetensors", + "model.layers.19.self_attn.o_proj.weight_scale": "model-00014-of-00086.safetensors", + "model.layers.19.self_attn.q_proj.weight": "model-00014-of-00086.safetensors", + "model.layers.19.self_attn.q_proj.weight_scale": "model-00014-of-00086.safetensors", + "model.layers.19.self_attn.v_proj.weight": "model-00014-of-00086.safetensors", + "model.layers.19.self_attn.v_proj.weight_scale": "model-00014-of-00086.safetensors", + "model.layers.2.input_layernorm.weight": "model-00003-of-00086.safetensors", + "model.layers.2.mlp.down_proj.weight": "model-00003-of-00086.safetensors", + "model.layers.2.mlp.down_proj.weight_scale": "model-00003-of-00086.safetensors", + "model.layers.2.mlp.gate_proj.weight": "model-00003-of-00086.safetensors", + "model.layers.2.mlp.gate_proj.weight_scale": "model-00003-of-00086.safetensors", + "model.layers.2.mlp.up_proj.weight": "model-00003-of-00086.safetensors", + "model.layers.2.mlp.up_proj.weight_scale": "model-00003-of-00086.safetensors", + "model.layers.2.post_attention_layernorm.weight": "model-00003-of-00086.safetensors", + "model.layers.2.self_attn.k_proj.weight": "model-00003-of-00086.safetensors", + "model.layers.2.self_attn.k_proj.weight_scale": "model-00003-of-00086.safetensors", + "model.layers.2.self_attn.o_proj.weight": "model-00003-of-00086.safetensors", + "model.layers.2.self_attn.o_proj.weight_scale": "model-00003-of-00086.safetensors", + "model.layers.2.self_attn.q_proj.weight": "model-00003-of-00086.safetensors", + "model.layers.2.self_attn.q_proj.weight_scale": "model-00003-of-00086.safetensors", + "model.layers.2.self_attn.v_proj.weight": "model-00003-of-00086.safetensors", + "model.layers.2.self_attn.v_proj.weight_scale": "model-00003-of-00086.safetensors", + "model.layers.20.input_layernorm.weight": "model-00015-of-00086.safetensors", + "model.layers.20.mlp.down_proj.weight": "model-00015-of-00086.safetensors", + "model.layers.20.mlp.down_proj.weight_scale": "model-00015-of-00086.safetensors", + "model.layers.20.mlp.gate_proj.weight": "model-00015-of-00086.safetensors", + "model.layers.20.mlp.gate_proj.weight_scale": "model-00015-of-00086.safetensors", + "model.layers.20.mlp.up_proj.weight": "model-00015-of-00086.safetensors", + "model.layers.20.mlp.up_proj.weight_scale": "model-00015-of-00086.safetensors", + "model.layers.20.post_attention_layernorm.weight": "model-00015-of-00086.safetensors", + "model.layers.20.self_attn.k_proj.weight": "model-00015-of-00086.safetensors", + "model.layers.20.self_attn.k_proj.weight_scale": "model-00015-of-00086.safetensors", + "model.layers.20.self_attn.o_proj.weight": "model-00015-of-00086.safetensors", + "model.layers.20.self_attn.o_proj.weight_scale": "model-00015-of-00086.safetensors", + "model.layers.20.self_attn.q_proj.weight": "model-00015-of-00086.safetensors", + "model.layers.20.self_attn.q_proj.weight_scale": "model-00015-of-00086.safetensors", + "model.layers.20.self_attn.v_proj.weight": "model-00015-of-00086.safetensors", + "model.layers.20.self_attn.v_proj.weight_scale": "model-00015-of-00086.safetensors", + "model.layers.21.input_layernorm.weight": "model-00016-of-00086.safetensors", + "model.layers.21.mlp.down_proj.weight": "model-00016-of-00086.safetensors", + "model.layers.21.mlp.down_proj.weight_scale": "model-00016-of-00086.safetensors", + "model.layers.21.mlp.gate_proj.weight": "model-00016-of-00086.safetensors", + "model.layers.21.mlp.gate_proj.weight_scale": "model-00016-of-00086.safetensors", + "model.layers.21.mlp.up_proj.weight": "model-00016-of-00086.safetensors", + "model.layers.21.mlp.up_proj.weight_scale": "model-00016-of-00086.safetensors", + "model.layers.21.post_attention_layernorm.weight": "model-00016-of-00086.safetensors", + "model.layers.21.self_attn.k_proj.weight": "model-00015-of-00086.safetensors", + "model.layers.21.self_attn.k_proj.weight_scale": "model-00015-of-00086.safetensors", + "model.layers.21.self_attn.o_proj.weight": "model-00015-of-00086.safetensors", + "model.layers.21.self_attn.o_proj.weight_scale": "model-00015-of-00086.safetensors", + "model.layers.21.self_attn.q_proj.weight": "model-00015-of-00086.safetensors", + "model.layers.21.self_attn.q_proj.weight_scale": "model-00015-of-00086.safetensors", + "model.layers.21.self_attn.v_proj.weight": "model-00015-of-00086.safetensors", + "model.layers.21.self_attn.v_proj.weight_scale": "model-00015-of-00086.safetensors", + "model.layers.22.input_layernorm.weight": "model-00017-of-00086.safetensors", + "model.layers.22.mlp.down_proj.weight": "model-00017-of-00086.safetensors", + "model.layers.22.mlp.down_proj.weight_scale": "model-00017-of-00086.safetensors", + "model.layers.22.mlp.gate_proj.weight": "model-00016-of-00086.safetensors", + "model.layers.22.mlp.gate_proj.weight_scale": "model-00016-of-00086.safetensors", + "model.layers.22.mlp.up_proj.weight": "model-00016-of-00086.safetensors", + "model.layers.22.mlp.up_proj.weight_scale": "model-00016-of-00086.safetensors", + "model.layers.22.post_attention_layernorm.weight": "model-00017-of-00086.safetensors", + "model.layers.22.self_attn.k_proj.weight": "model-00016-of-00086.safetensors", + "model.layers.22.self_attn.k_proj.weight_scale": "model-00016-of-00086.safetensors", + "model.layers.22.self_attn.o_proj.weight": "model-00016-of-00086.safetensors", + "model.layers.22.self_attn.o_proj.weight_scale": "model-00016-of-00086.safetensors", + "model.layers.22.self_attn.q_proj.weight": "model-00016-of-00086.safetensors", + "model.layers.22.self_attn.q_proj.weight_scale": "model-00016-of-00086.safetensors", + "model.layers.22.self_attn.v_proj.weight": "model-00016-of-00086.safetensors", + "model.layers.22.self_attn.v_proj.weight_scale": "model-00016-of-00086.safetensors", + "model.layers.23.input_layernorm.weight": "model-00017-of-00086.safetensors", + "model.layers.23.mlp.down_proj.weight": "model-00017-of-00086.safetensors", + "model.layers.23.mlp.down_proj.weight_scale": "model-00017-of-00086.safetensors", + "model.layers.23.mlp.gate_proj.weight": "model-00017-of-00086.safetensors", + "model.layers.23.mlp.gate_proj.weight_scale": "model-00017-of-00086.safetensors", + "model.layers.23.mlp.up_proj.weight": "model-00017-of-00086.safetensors", + "model.layers.23.mlp.up_proj.weight_scale": "model-00017-of-00086.safetensors", + "model.layers.23.post_attention_layernorm.weight": "model-00017-of-00086.safetensors", + "model.layers.23.self_attn.k_proj.weight": "model-00017-of-00086.safetensors", + "model.layers.23.self_attn.k_proj.weight_scale": "model-00017-of-00086.safetensors", + "model.layers.23.self_attn.o_proj.weight": "model-00017-of-00086.safetensors", + "model.layers.23.self_attn.o_proj.weight_scale": "model-00017-of-00086.safetensors", + "model.layers.23.self_attn.q_proj.weight": "model-00017-of-00086.safetensors", + "model.layers.23.self_attn.q_proj.weight_scale": "model-00017-of-00086.safetensors", + "model.layers.23.self_attn.v_proj.weight": "model-00017-of-00086.safetensors", + "model.layers.23.self_attn.v_proj.weight_scale": "model-00017-of-00086.safetensors", + "model.layers.24.input_layernorm.weight": "model-00018-of-00086.safetensors", + "model.layers.24.mlp.down_proj.weight": "model-00018-of-00086.safetensors", + "model.layers.24.mlp.down_proj.weight_scale": "model-00018-of-00086.safetensors", + "model.layers.24.mlp.gate_proj.weight": "model-00018-of-00086.safetensors", + "model.layers.24.mlp.gate_proj.weight_scale": "model-00018-of-00086.safetensors", + "model.layers.24.mlp.up_proj.weight": "model-00018-of-00086.safetensors", + "model.layers.24.mlp.up_proj.weight_scale": "model-00018-of-00086.safetensors", + "model.layers.24.post_attention_layernorm.weight": "model-00018-of-00086.safetensors", + "model.layers.24.self_attn.k_proj.weight": "model-00017-of-00086.safetensors", + "model.layers.24.self_attn.k_proj.weight_scale": "model-00017-of-00086.safetensors", + "model.layers.24.self_attn.o_proj.weight": "model-00017-of-00086.safetensors", + "model.layers.24.self_attn.o_proj.weight_scale": "model-00017-of-00086.safetensors", + "model.layers.24.self_attn.q_proj.weight": "model-00017-of-00086.safetensors", + "model.layers.24.self_attn.q_proj.weight_scale": "model-00017-of-00086.safetensors", + "model.layers.24.self_attn.v_proj.weight": "model-00017-of-00086.safetensors", + "model.layers.24.self_attn.v_proj.weight_scale": "model-00017-of-00086.safetensors", + "model.layers.25.input_layernorm.weight": "model-00019-of-00086.safetensors", + "model.layers.25.mlp.down_proj.weight": "model-00019-of-00086.safetensors", + "model.layers.25.mlp.down_proj.weight_scale": "model-00019-of-00086.safetensors", + "model.layers.25.mlp.gate_proj.weight": "model-00018-of-00086.safetensors", + "model.layers.25.mlp.gate_proj.weight_scale": "model-00018-of-00086.safetensors", + "model.layers.25.mlp.up_proj.weight": "model-00018-of-00086.safetensors", + "model.layers.25.mlp.up_proj.weight_scale": "model-00018-of-00086.safetensors", + "model.layers.25.post_attention_layernorm.weight": "model-00019-of-00086.safetensors", + "model.layers.25.self_attn.k_proj.weight": "model-00018-of-00086.safetensors", + "model.layers.25.self_attn.k_proj.weight_scale": "model-00018-of-00086.safetensors", + "model.layers.25.self_attn.o_proj.weight": "model-00018-of-00086.safetensors", + "model.layers.25.self_attn.o_proj.weight_scale": "model-00018-of-00086.safetensors", + "model.layers.25.self_attn.q_proj.weight": "model-00018-of-00086.safetensors", + "model.layers.25.self_attn.q_proj.weight_scale": "model-00018-of-00086.safetensors", + "model.layers.25.self_attn.v_proj.weight": "model-00018-of-00086.safetensors", + "model.layers.25.self_attn.v_proj.weight_scale": "model-00018-of-00086.safetensors", + "model.layers.26.input_layernorm.weight": "model-00019-of-00086.safetensors", + "model.layers.26.mlp.down_proj.weight": "model-00019-of-00086.safetensors", + "model.layers.26.mlp.down_proj.weight_scale": "model-00019-of-00086.safetensors", + "model.layers.26.mlp.gate_proj.weight": "model-00019-of-00086.safetensors", + "model.layers.26.mlp.gate_proj.weight_scale": "model-00019-of-00086.safetensors", + "model.layers.26.mlp.up_proj.weight": "model-00019-of-00086.safetensors", + "model.layers.26.mlp.up_proj.weight_scale": "model-00019-of-00086.safetensors", + "model.layers.26.post_attention_layernorm.weight": "model-00019-of-00086.safetensors", + "model.layers.26.self_attn.k_proj.weight": "model-00019-of-00086.safetensors", + "model.layers.26.self_attn.k_proj.weight_scale": "model-00019-of-00086.safetensors", + "model.layers.26.self_attn.o_proj.weight": "model-00019-of-00086.safetensors", + "model.layers.26.self_attn.o_proj.weight_scale": "model-00019-of-00086.safetensors", + "model.layers.26.self_attn.q_proj.weight": "model-00019-of-00086.safetensors", + "model.layers.26.self_attn.q_proj.weight_scale": "model-00019-of-00086.safetensors", + "model.layers.26.self_attn.v_proj.weight": "model-00019-of-00086.safetensors", + "model.layers.26.self_attn.v_proj.weight_scale": "model-00019-of-00086.safetensors", + "model.layers.27.input_layernorm.weight": "model-00020-of-00086.safetensors", + "model.layers.27.mlp.down_proj.weight": "model-00020-of-00086.safetensors", + "model.layers.27.mlp.down_proj.weight_scale": "model-00020-of-00086.safetensors", + "model.layers.27.mlp.gate_proj.weight": "model-00020-of-00086.safetensors", + "model.layers.27.mlp.gate_proj.weight_scale": "model-00020-of-00086.safetensors", + "model.layers.27.mlp.up_proj.weight": "model-00020-of-00086.safetensors", + "model.layers.27.mlp.up_proj.weight_scale": "model-00020-of-00086.safetensors", + "model.layers.27.post_attention_layernorm.weight": "model-00020-of-00086.safetensors", + "model.layers.27.self_attn.k_proj.weight": "model-00019-of-00086.safetensors", + "model.layers.27.self_attn.k_proj.weight_scale": "model-00019-of-00086.safetensors", + "model.layers.27.self_attn.o_proj.weight": "model-00019-of-00086.safetensors", + "model.layers.27.self_attn.o_proj.weight_scale": "model-00019-of-00086.safetensors", + "model.layers.27.self_attn.q_proj.weight": "model-00019-of-00086.safetensors", + "model.layers.27.self_attn.q_proj.weight_scale": "model-00019-of-00086.safetensors", + "model.layers.27.self_attn.v_proj.weight": "model-00019-of-00086.safetensors", + "model.layers.27.self_attn.v_proj.weight_scale": "model-00019-of-00086.safetensors", + "model.layers.28.input_layernorm.weight": "model-00021-of-00086.safetensors", + "model.layers.28.mlp.down_proj.weight": "model-00021-of-00086.safetensors", + "model.layers.28.mlp.down_proj.weight_scale": "model-00021-of-00086.safetensors", + "model.layers.28.mlp.gate_proj.weight": "model-00020-of-00086.safetensors", + "model.layers.28.mlp.gate_proj.weight_scale": "model-00020-of-00086.safetensors", + "model.layers.28.mlp.up_proj.weight": "model-00020-of-00086.safetensors", + "model.layers.28.mlp.up_proj.weight_scale": "model-00020-of-00086.safetensors", + "model.layers.28.post_attention_layernorm.weight": "model-00021-of-00086.safetensors", + "model.layers.28.self_attn.k_proj.weight": "model-00020-of-00086.safetensors", + "model.layers.28.self_attn.k_proj.weight_scale": "model-00020-of-00086.safetensors", + "model.layers.28.self_attn.o_proj.weight": "model-00020-of-00086.safetensors", + "model.layers.28.self_attn.o_proj.weight_scale": "model-00020-of-00086.safetensors", + "model.layers.28.self_attn.q_proj.weight": "model-00020-of-00086.safetensors", + "model.layers.28.self_attn.q_proj.weight_scale": "model-00020-of-00086.safetensors", + "model.layers.28.self_attn.v_proj.weight": "model-00020-of-00086.safetensors", + "model.layers.28.self_attn.v_proj.weight_scale": "model-00020-of-00086.safetensors", + "model.layers.29.input_layernorm.weight": "model-00021-of-00086.safetensors", + "model.layers.29.mlp.down_proj.weight": "model-00021-of-00086.safetensors", + "model.layers.29.mlp.down_proj.weight_scale": "model-00021-of-00086.safetensors", + "model.layers.29.mlp.gate_proj.weight": "model-00021-of-00086.safetensors", + "model.layers.29.mlp.gate_proj.weight_scale": "model-00021-of-00086.safetensors", + "model.layers.29.mlp.up_proj.weight": "model-00021-of-00086.safetensors", + "model.layers.29.mlp.up_proj.weight_scale": "model-00021-of-00086.safetensors", + "model.layers.29.post_attention_layernorm.weight": "model-00021-of-00086.safetensors", + "model.layers.29.self_attn.k_proj.weight": "model-00021-of-00086.safetensors", + "model.layers.29.self_attn.k_proj.weight_scale": "model-00021-of-00086.safetensors", + "model.layers.29.self_attn.o_proj.weight": "model-00021-of-00086.safetensors", + "model.layers.29.self_attn.o_proj.weight_scale": "model-00021-of-00086.safetensors", + "model.layers.29.self_attn.q_proj.weight": "model-00021-of-00086.safetensors", + "model.layers.29.self_attn.q_proj.weight_scale": "model-00021-of-00086.safetensors", + "model.layers.29.self_attn.v_proj.weight": "model-00021-of-00086.safetensors", + "model.layers.29.self_attn.v_proj.weight_scale": "model-00021-of-00086.safetensors", + "model.layers.3.input_layernorm.weight": "model-00004-of-00086.safetensors", + "model.layers.3.mlp.down_proj.weight": "model-00004-of-00086.safetensors", + "model.layers.3.mlp.down_proj.weight_scale": "model-00004-of-00086.safetensors", + "model.layers.3.mlp.gate_proj.weight": "model-00004-of-00086.safetensors", + "model.layers.3.mlp.gate_proj.weight_scale": "model-00004-of-00086.safetensors", + "model.layers.3.mlp.up_proj.weight": "model-00004-of-00086.safetensors", + "model.layers.3.mlp.up_proj.weight_scale": "model-00004-of-00086.safetensors", + "model.layers.3.post_attention_layernorm.weight": "model-00004-of-00086.safetensors", + "model.layers.3.self_attn.k_proj.weight": "model-00003-of-00086.safetensors", + "model.layers.3.self_attn.k_proj.weight_scale": "model-00003-of-00086.safetensors", + "model.layers.3.self_attn.o_proj.weight": "model-00003-of-00086.safetensors", + "model.layers.3.self_attn.o_proj.weight_scale": "model-00003-of-00086.safetensors", + "model.layers.3.self_attn.q_proj.weight": "model-00003-of-00086.safetensors", + "model.layers.3.self_attn.q_proj.weight_scale": "model-00003-of-00086.safetensors", + "model.layers.3.self_attn.v_proj.weight": "model-00003-of-00086.safetensors", + "model.layers.3.self_attn.v_proj.weight_scale": "model-00003-of-00086.safetensors", + "model.layers.30.input_layernorm.weight": "model-00022-of-00086.safetensors", + "model.layers.30.mlp.down_proj.weight": "model-00022-of-00086.safetensors", + "model.layers.30.mlp.down_proj.weight_scale": "model-00022-of-00086.safetensors", + "model.layers.30.mlp.gate_proj.weight": "model-00022-of-00086.safetensors", + "model.layers.30.mlp.gate_proj.weight_scale": "model-00022-of-00086.safetensors", + "model.layers.30.mlp.up_proj.weight": "model-00022-of-00086.safetensors", + "model.layers.30.mlp.up_proj.weight_scale": "model-00022-of-00086.safetensors", + "model.layers.30.post_attention_layernorm.weight": "model-00022-of-00086.safetensors", + "model.layers.30.self_attn.k_proj.weight": "model-00021-of-00086.safetensors", + "model.layers.30.self_attn.k_proj.weight_scale": "model-00021-of-00086.safetensors", + "model.layers.30.self_attn.o_proj.weight": "model-00021-of-00086.safetensors", + "model.layers.30.self_attn.o_proj.weight_scale": "model-00021-of-00086.safetensors", + "model.layers.30.self_attn.q_proj.weight": "model-00021-of-00086.safetensors", + "model.layers.30.self_attn.q_proj.weight_scale": "model-00021-of-00086.safetensors", + "model.layers.30.self_attn.v_proj.weight": "model-00021-of-00086.safetensors", + "model.layers.30.self_attn.v_proj.weight_scale": "model-00021-of-00086.safetensors", + "model.layers.31.input_layernorm.weight": "model-00023-of-00086.safetensors", + "model.layers.31.mlp.down_proj.weight": "model-00023-of-00086.safetensors", + "model.layers.31.mlp.down_proj.weight_scale": "model-00023-of-00086.safetensors", + "model.layers.31.mlp.gate_proj.weight": "model-00022-of-00086.safetensors", + "model.layers.31.mlp.gate_proj.weight_scale": "model-00022-of-00086.safetensors", + "model.layers.31.mlp.up_proj.weight": "model-00022-of-00086.safetensors", + "model.layers.31.mlp.up_proj.weight_scale": "model-00022-of-00086.safetensors", + "model.layers.31.post_attention_layernorm.weight": "model-00023-of-00086.safetensors", + "model.layers.31.self_attn.k_proj.weight": "model-00022-of-00086.safetensors", + "model.layers.31.self_attn.k_proj.weight_scale": "model-00022-of-00086.safetensors", + "model.layers.31.self_attn.o_proj.weight": "model-00022-of-00086.safetensors", + "model.layers.31.self_attn.o_proj.weight_scale": "model-00022-of-00086.safetensors", + "model.layers.31.self_attn.q_proj.weight": "model-00022-of-00086.safetensors", + "model.layers.31.self_attn.q_proj.weight_scale": "model-00022-of-00086.safetensors", + "model.layers.31.self_attn.v_proj.weight": "model-00022-of-00086.safetensors", + "model.layers.31.self_attn.v_proj.weight_scale": "model-00022-of-00086.safetensors", + "model.layers.32.input_layernorm.weight": "model-00023-of-00086.safetensors", + "model.layers.32.mlp.down_proj.weight": "model-00023-of-00086.safetensors", + "model.layers.32.mlp.down_proj.weight_scale": "model-00023-of-00086.safetensors", + "model.layers.32.mlp.gate_proj.weight": "model-00023-of-00086.safetensors", + "model.layers.32.mlp.gate_proj.weight_scale": "model-00023-of-00086.safetensors", + "model.layers.32.mlp.up_proj.weight": "model-00023-of-00086.safetensors", + "model.layers.32.mlp.up_proj.weight_scale": "model-00023-of-00086.safetensors", + "model.layers.32.post_attention_layernorm.weight": "model-00023-of-00086.safetensors", + "model.layers.32.self_attn.k_proj.weight": "model-00023-of-00086.safetensors", + "model.layers.32.self_attn.k_proj.weight_scale": "model-00023-of-00086.safetensors", + "model.layers.32.self_attn.o_proj.weight": "model-00023-of-00086.safetensors", + "model.layers.32.self_attn.o_proj.weight_scale": "model-00023-of-00086.safetensors", + "model.layers.32.self_attn.q_proj.weight": "model-00023-of-00086.safetensors", + "model.layers.32.self_attn.q_proj.weight_scale": "model-00023-of-00086.safetensors", + "model.layers.32.self_attn.v_proj.weight": "model-00023-of-00086.safetensors", + "model.layers.32.self_attn.v_proj.weight_scale": "model-00023-of-00086.safetensors", + "model.layers.33.input_layernorm.weight": "model-00024-of-00086.safetensors", + "model.layers.33.mlp.down_proj.weight": "model-00024-of-00086.safetensors", + "model.layers.33.mlp.down_proj.weight_scale": "model-00024-of-00086.safetensors", + "model.layers.33.mlp.gate_proj.weight": "model-00024-of-00086.safetensors", + "model.layers.33.mlp.gate_proj.weight_scale": "model-00024-of-00086.safetensors", + "model.layers.33.mlp.up_proj.weight": "model-00024-of-00086.safetensors", + "model.layers.33.mlp.up_proj.weight_scale": "model-00024-of-00086.safetensors", + "model.layers.33.post_attention_layernorm.weight": "model-00024-of-00086.safetensors", + "model.layers.33.self_attn.k_proj.weight": "model-00023-of-00086.safetensors", + "model.layers.33.self_attn.k_proj.weight_scale": "model-00023-of-00086.safetensors", + "model.layers.33.self_attn.o_proj.weight": "model-00023-of-00086.safetensors", + "model.layers.33.self_attn.o_proj.weight_scale": "model-00023-of-00086.safetensors", + "model.layers.33.self_attn.q_proj.weight": "model-00023-of-00086.safetensors", + "model.layers.33.self_attn.q_proj.weight_scale": "model-00023-of-00086.safetensors", + "model.layers.33.self_attn.v_proj.weight": "model-00023-of-00086.safetensors", + "model.layers.33.self_attn.v_proj.weight_scale": "model-00023-of-00086.safetensors", + "model.layers.34.input_layernorm.weight": "model-00025-of-00086.safetensors", + "model.layers.34.mlp.down_proj.weight": "model-00025-of-00086.safetensors", + "model.layers.34.mlp.down_proj.weight_scale": "model-00025-of-00086.safetensors", + "model.layers.34.mlp.gate_proj.weight": "model-00024-of-00086.safetensors", + "model.layers.34.mlp.gate_proj.weight_scale": "model-00024-of-00086.safetensors", + "model.layers.34.mlp.up_proj.weight": "model-00024-of-00086.safetensors", + "model.layers.34.mlp.up_proj.weight_scale": "model-00024-of-00086.safetensors", + "model.layers.34.post_attention_layernorm.weight": "model-00025-of-00086.safetensors", + "model.layers.34.self_attn.k_proj.weight": "model-00024-of-00086.safetensors", + "model.layers.34.self_attn.k_proj.weight_scale": "model-00024-of-00086.safetensors", + "model.layers.34.self_attn.o_proj.weight": "model-00024-of-00086.safetensors", + "model.layers.34.self_attn.o_proj.weight_scale": "model-00024-of-00086.safetensors", + "model.layers.34.self_attn.q_proj.weight": "model-00024-of-00086.safetensors", + "model.layers.34.self_attn.q_proj.weight_scale": "model-00024-of-00086.safetensors", + "model.layers.34.self_attn.v_proj.weight": "model-00024-of-00086.safetensors", + "model.layers.34.self_attn.v_proj.weight_scale": "model-00024-of-00086.safetensors", + "model.layers.35.input_layernorm.weight": "model-00025-of-00086.safetensors", + "model.layers.35.mlp.down_proj.weight": "model-00025-of-00086.safetensors", + "model.layers.35.mlp.down_proj.weight_scale": "model-00025-of-00086.safetensors", + "model.layers.35.mlp.gate_proj.weight": "model-00025-of-00086.safetensors", + "model.layers.35.mlp.gate_proj.weight_scale": "model-00025-of-00086.safetensors", + "model.layers.35.mlp.up_proj.weight": "model-00025-of-00086.safetensors", + "model.layers.35.mlp.up_proj.weight_scale": "model-00025-of-00086.safetensors", + "model.layers.35.post_attention_layernorm.weight": "model-00025-of-00086.safetensors", + "model.layers.35.self_attn.k_proj.weight": "model-00025-of-00086.safetensors", + "model.layers.35.self_attn.k_proj.weight_scale": "model-00025-of-00086.safetensors", + "model.layers.35.self_attn.o_proj.weight": "model-00025-of-00086.safetensors", + "model.layers.35.self_attn.o_proj.weight_scale": "model-00025-of-00086.safetensors", + "model.layers.35.self_attn.q_proj.weight": "model-00025-of-00086.safetensors", + "model.layers.35.self_attn.q_proj.weight_scale": "model-00025-of-00086.safetensors", + "model.layers.35.self_attn.v_proj.weight": "model-00025-of-00086.safetensors", + "model.layers.35.self_attn.v_proj.weight_scale": "model-00025-of-00086.safetensors", + "model.layers.36.input_layernorm.weight": "model-00026-of-00086.safetensors", + "model.layers.36.mlp.down_proj.weight": "model-00026-of-00086.safetensors", + "model.layers.36.mlp.down_proj.weight_scale": "model-00026-of-00086.safetensors", + "model.layers.36.mlp.gate_proj.weight": "model-00026-of-00086.safetensors", + "model.layers.36.mlp.gate_proj.weight_scale": "model-00026-of-00086.safetensors", + "model.layers.36.mlp.up_proj.weight": "model-00026-of-00086.safetensors", + "model.layers.36.mlp.up_proj.weight_scale": "model-00026-of-00086.safetensors", + "model.layers.36.post_attention_layernorm.weight": "model-00026-of-00086.safetensors", + "model.layers.36.self_attn.k_proj.weight": "model-00025-of-00086.safetensors", + "model.layers.36.self_attn.k_proj.weight_scale": "model-00025-of-00086.safetensors", + "model.layers.36.self_attn.o_proj.weight": "model-00025-of-00086.safetensors", + "model.layers.36.self_attn.o_proj.weight_scale": "model-00025-of-00086.safetensors", + "model.layers.36.self_attn.q_proj.weight": "model-00025-of-00086.safetensors", + "model.layers.36.self_attn.q_proj.weight_scale": "model-00025-of-00086.safetensors", + "model.layers.36.self_attn.v_proj.weight": "model-00025-of-00086.safetensors", + "model.layers.36.self_attn.v_proj.weight_scale": "model-00025-of-00086.safetensors", + "model.layers.37.input_layernorm.weight": "model-00027-of-00086.safetensors", + "model.layers.37.mlp.down_proj.weight": "model-00027-of-00086.safetensors", + "model.layers.37.mlp.down_proj.weight_scale": "model-00027-of-00086.safetensors", + "model.layers.37.mlp.gate_proj.weight": "model-00026-of-00086.safetensors", + "model.layers.37.mlp.gate_proj.weight_scale": "model-00026-of-00086.safetensors", + "model.layers.37.mlp.up_proj.weight": "model-00026-of-00086.safetensors", + "model.layers.37.mlp.up_proj.weight_scale": "model-00026-of-00086.safetensors", + "model.layers.37.post_attention_layernorm.weight": "model-00027-of-00086.safetensors", + "model.layers.37.self_attn.k_proj.weight": "model-00026-of-00086.safetensors", + "model.layers.37.self_attn.k_proj.weight_scale": "model-00026-of-00086.safetensors", + "model.layers.37.self_attn.o_proj.weight": "model-00026-of-00086.safetensors", + "model.layers.37.self_attn.o_proj.weight_scale": "model-00026-of-00086.safetensors", + "model.layers.37.self_attn.q_proj.weight": "model-00026-of-00086.safetensors", + "model.layers.37.self_attn.q_proj.weight_scale": "model-00026-of-00086.safetensors", + "model.layers.37.self_attn.v_proj.weight": "model-00026-of-00086.safetensors", + "model.layers.37.self_attn.v_proj.weight_scale": "model-00026-of-00086.safetensors", + "model.layers.38.input_layernorm.weight": "model-00027-of-00086.safetensors", + "model.layers.38.mlp.down_proj.weight": "model-00027-of-00086.safetensors", + "model.layers.38.mlp.down_proj.weight_scale": "model-00027-of-00086.safetensors", + "model.layers.38.mlp.gate_proj.weight": "model-00027-of-00086.safetensors", + "model.layers.38.mlp.gate_proj.weight_scale": "model-00027-of-00086.safetensors", + "model.layers.38.mlp.up_proj.weight": "model-00027-of-00086.safetensors", + "model.layers.38.mlp.up_proj.weight_scale": "model-00027-of-00086.safetensors", + "model.layers.38.post_attention_layernorm.weight": "model-00027-of-00086.safetensors", + "model.layers.38.self_attn.k_proj.weight": "model-00027-of-00086.safetensors", + "model.layers.38.self_attn.k_proj.weight_scale": "model-00027-of-00086.safetensors", + "model.layers.38.self_attn.o_proj.weight": "model-00027-of-00086.safetensors", + "model.layers.38.self_attn.o_proj.weight_scale": "model-00027-of-00086.safetensors", + "model.layers.38.self_attn.q_proj.weight": "model-00027-of-00086.safetensors", + "model.layers.38.self_attn.q_proj.weight_scale": "model-00027-of-00086.safetensors", + "model.layers.38.self_attn.v_proj.weight": "model-00027-of-00086.safetensors", + "model.layers.38.self_attn.v_proj.weight_scale": "model-00027-of-00086.safetensors", + "model.layers.39.input_layernorm.weight": "model-00028-of-00086.safetensors", + "model.layers.39.mlp.down_proj.weight": "model-00028-of-00086.safetensors", + "model.layers.39.mlp.down_proj.weight_scale": "model-00028-of-00086.safetensors", + "model.layers.39.mlp.gate_proj.weight": "model-00028-of-00086.safetensors", + "model.layers.39.mlp.gate_proj.weight_scale": "model-00028-of-00086.safetensors", + "model.layers.39.mlp.up_proj.weight": "model-00028-of-00086.safetensors", + "model.layers.39.mlp.up_proj.weight_scale": "model-00028-of-00086.safetensors", + "model.layers.39.post_attention_layernorm.weight": "model-00028-of-00086.safetensors", + "model.layers.39.self_attn.k_proj.weight": "model-00027-of-00086.safetensors", + "model.layers.39.self_attn.k_proj.weight_scale": "model-00027-of-00086.safetensors", + "model.layers.39.self_attn.o_proj.weight": "model-00027-of-00086.safetensors", + "model.layers.39.self_attn.o_proj.weight_scale": "model-00027-of-00086.safetensors", + "model.layers.39.self_attn.q_proj.weight": "model-00027-of-00086.safetensors", + "model.layers.39.self_attn.q_proj.weight_scale": "model-00027-of-00086.safetensors", + "model.layers.39.self_attn.v_proj.weight": "model-00027-of-00086.safetensors", + "model.layers.39.self_attn.v_proj.weight_scale": "model-00027-of-00086.safetensors", + "model.layers.4.input_layernorm.weight": "model-00005-of-00086.safetensors", + "model.layers.4.mlp.down_proj.weight": "model-00005-of-00086.safetensors", + "model.layers.4.mlp.down_proj.weight_scale": "model-00005-of-00086.safetensors", + "model.layers.4.mlp.gate_proj.weight": "model-00004-of-00086.safetensors", + "model.layers.4.mlp.gate_proj.weight_scale": "model-00004-of-00086.safetensors", + "model.layers.4.mlp.up_proj.weight": "model-00004-of-00086.safetensors", + "model.layers.4.mlp.up_proj.weight_scale": "model-00004-of-00086.safetensors", + "model.layers.4.post_attention_layernorm.weight": "model-00005-of-00086.safetensors", + "model.layers.4.self_attn.k_proj.weight": "model-00004-of-00086.safetensors", + "model.layers.4.self_attn.k_proj.weight_scale": "model-00004-of-00086.safetensors", + "model.layers.4.self_attn.o_proj.weight": "model-00004-of-00086.safetensors", + "model.layers.4.self_attn.o_proj.weight_scale": "model-00004-of-00086.safetensors", + "model.layers.4.self_attn.q_proj.weight": "model-00004-of-00086.safetensors", + "model.layers.4.self_attn.q_proj.weight_scale": "model-00004-of-00086.safetensors", + "model.layers.4.self_attn.v_proj.weight": "model-00004-of-00086.safetensors", + "model.layers.4.self_attn.v_proj.weight_scale": "model-00004-of-00086.safetensors", + "model.layers.40.input_layernorm.weight": "model-00029-of-00086.safetensors", + "model.layers.40.mlp.down_proj.weight": "model-00029-of-00086.safetensors", + "model.layers.40.mlp.down_proj.weight_scale": "model-00029-of-00086.safetensors", + "model.layers.40.mlp.gate_proj.weight": "model-00028-of-00086.safetensors", + "model.layers.40.mlp.gate_proj.weight_scale": "model-00028-of-00086.safetensors", + "model.layers.40.mlp.up_proj.weight": "model-00028-of-00086.safetensors", + "model.layers.40.mlp.up_proj.weight_scale": "model-00028-of-00086.safetensors", + "model.layers.40.post_attention_layernorm.weight": "model-00029-of-00086.safetensors", + "model.layers.40.self_attn.k_proj.weight": "model-00028-of-00086.safetensors", + "model.layers.40.self_attn.k_proj.weight_scale": "model-00028-of-00086.safetensors", + "model.layers.40.self_attn.o_proj.weight": "model-00028-of-00086.safetensors", + "model.layers.40.self_attn.o_proj.weight_scale": "model-00028-of-00086.safetensors", + "model.layers.40.self_attn.q_proj.weight": "model-00028-of-00086.safetensors", + "model.layers.40.self_attn.q_proj.weight_scale": "model-00028-of-00086.safetensors", + "model.layers.40.self_attn.v_proj.weight": "model-00028-of-00086.safetensors", + "model.layers.40.self_attn.v_proj.weight_scale": "model-00028-of-00086.safetensors", + "model.layers.41.input_layernorm.weight": "model-00029-of-00086.safetensors", + "model.layers.41.mlp.down_proj.weight": "model-00029-of-00086.safetensors", + "model.layers.41.mlp.down_proj.weight_scale": "model-00029-of-00086.safetensors", + "model.layers.41.mlp.gate_proj.weight": "model-00029-of-00086.safetensors", + "model.layers.41.mlp.gate_proj.weight_scale": "model-00029-of-00086.safetensors", + "model.layers.41.mlp.up_proj.weight": "model-00029-of-00086.safetensors", + "model.layers.41.mlp.up_proj.weight_scale": "model-00029-of-00086.safetensors", + "model.layers.41.post_attention_layernorm.weight": "model-00029-of-00086.safetensors", + "model.layers.41.self_attn.k_proj.weight": "model-00029-of-00086.safetensors", + "model.layers.41.self_attn.k_proj.weight_scale": "model-00029-of-00086.safetensors", + "model.layers.41.self_attn.o_proj.weight": "model-00029-of-00086.safetensors", + "model.layers.41.self_attn.o_proj.weight_scale": "model-00029-of-00086.safetensors", + "model.layers.41.self_attn.q_proj.weight": "model-00029-of-00086.safetensors", + "model.layers.41.self_attn.q_proj.weight_scale": "model-00029-of-00086.safetensors", + "model.layers.41.self_attn.v_proj.weight": "model-00029-of-00086.safetensors", + "model.layers.41.self_attn.v_proj.weight_scale": "model-00029-of-00086.safetensors", + "model.layers.42.input_layernorm.weight": "model-00030-of-00086.safetensors", + "model.layers.42.mlp.down_proj.weight": "model-00030-of-00086.safetensors", + "model.layers.42.mlp.down_proj.weight_scale": "model-00030-of-00086.safetensors", + "model.layers.42.mlp.gate_proj.weight": "model-00030-of-00086.safetensors", + "model.layers.42.mlp.gate_proj.weight_scale": "model-00030-of-00086.safetensors", + "model.layers.42.mlp.up_proj.weight": "model-00030-of-00086.safetensors", + "model.layers.42.mlp.up_proj.weight_scale": "model-00030-of-00086.safetensors", + "model.layers.42.post_attention_layernorm.weight": "model-00030-of-00086.safetensors", + "model.layers.42.self_attn.k_proj.weight": "model-00029-of-00086.safetensors", + "model.layers.42.self_attn.k_proj.weight_scale": "model-00029-of-00086.safetensors", + "model.layers.42.self_attn.o_proj.weight": "model-00029-of-00086.safetensors", + "model.layers.42.self_attn.o_proj.weight_scale": "model-00029-of-00086.safetensors", + "model.layers.42.self_attn.q_proj.weight": "model-00029-of-00086.safetensors", + "model.layers.42.self_attn.q_proj.weight_scale": "model-00029-of-00086.safetensors", + "model.layers.42.self_attn.v_proj.weight": "model-00029-of-00086.safetensors", + "model.layers.42.self_attn.v_proj.weight_scale": "model-00029-of-00086.safetensors", + "model.layers.43.input_layernorm.weight": "model-00031-of-00086.safetensors", + "model.layers.43.mlp.down_proj.weight": "model-00031-of-00086.safetensors", + "model.layers.43.mlp.down_proj.weight_scale": "model-00031-of-00086.safetensors", + "model.layers.43.mlp.gate_proj.weight": "model-00030-of-00086.safetensors", + "model.layers.43.mlp.gate_proj.weight_scale": "model-00030-of-00086.safetensors", + "model.layers.43.mlp.up_proj.weight": "model-00030-of-00086.safetensors", + "model.layers.43.mlp.up_proj.weight_scale": "model-00030-of-00086.safetensors", + "model.layers.43.post_attention_layernorm.weight": "model-00031-of-00086.safetensors", + "model.layers.43.self_attn.k_proj.weight": "model-00030-of-00086.safetensors", + "model.layers.43.self_attn.k_proj.weight_scale": "model-00030-of-00086.safetensors", + "model.layers.43.self_attn.o_proj.weight": "model-00030-of-00086.safetensors", + "model.layers.43.self_attn.o_proj.weight_scale": "model-00030-of-00086.safetensors", + "model.layers.43.self_attn.q_proj.weight": "model-00030-of-00086.safetensors", + "model.layers.43.self_attn.q_proj.weight_scale": "model-00030-of-00086.safetensors", + "model.layers.43.self_attn.v_proj.weight": "model-00030-of-00086.safetensors", + "model.layers.43.self_attn.v_proj.weight_scale": "model-00030-of-00086.safetensors", + "model.layers.44.input_layernorm.weight": "model-00031-of-00086.safetensors", + "model.layers.44.mlp.down_proj.weight": "model-00031-of-00086.safetensors", + "model.layers.44.mlp.down_proj.weight_scale": "model-00031-of-00086.safetensors", + "model.layers.44.mlp.gate_proj.weight": "model-00031-of-00086.safetensors", + "model.layers.44.mlp.gate_proj.weight_scale": "model-00031-of-00086.safetensors", + "model.layers.44.mlp.up_proj.weight": "model-00031-of-00086.safetensors", + "model.layers.44.mlp.up_proj.weight_scale": "model-00031-of-00086.safetensors", + "model.layers.44.post_attention_layernorm.weight": "model-00031-of-00086.safetensors", + "model.layers.44.self_attn.k_proj.weight": "model-00031-of-00086.safetensors", + "model.layers.44.self_attn.k_proj.weight_scale": "model-00031-of-00086.safetensors", + "model.layers.44.self_attn.o_proj.weight": "model-00031-of-00086.safetensors", + "model.layers.44.self_attn.o_proj.weight_scale": "model-00031-of-00086.safetensors", + "model.layers.44.self_attn.q_proj.weight": "model-00031-of-00086.safetensors", + "model.layers.44.self_attn.q_proj.weight_scale": "model-00031-of-00086.safetensors", + "model.layers.44.self_attn.v_proj.weight": "model-00031-of-00086.safetensors", + "model.layers.44.self_attn.v_proj.weight_scale": "model-00031-of-00086.safetensors", + "model.layers.45.input_layernorm.weight": "model-00032-of-00086.safetensors", + "model.layers.45.mlp.down_proj.weight": "model-00032-of-00086.safetensors", + "model.layers.45.mlp.down_proj.weight_scale": "model-00032-of-00086.safetensors", + "model.layers.45.mlp.gate_proj.weight": "model-00032-of-00086.safetensors", + "model.layers.45.mlp.gate_proj.weight_scale": "model-00032-of-00086.safetensors", + "model.layers.45.mlp.up_proj.weight": "model-00032-of-00086.safetensors", + "model.layers.45.mlp.up_proj.weight_scale": "model-00032-of-00086.safetensors", + "model.layers.45.post_attention_layernorm.weight": "model-00032-of-00086.safetensors", + "model.layers.45.self_attn.k_proj.weight": "model-00031-of-00086.safetensors", + "model.layers.45.self_attn.k_proj.weight_scale": "model-00031-of-00086.safetensors", + "model.layers.45.self_attn.o_proj.weight": "model-00031-of-00086.safetensors", + "model.layers.45.self_attn.o_proj.weight_scale": "model-00031-of-00086.safetensors", + "model.layers.45.self_attn.q_proj.weight": "model-00031-of-00086.safetensors", + "model.layers.45.self_attn.q_proj.weight_scale": "model-00031-of-00086.safetensors", + "model.layers.45.self_attn.v_proj.weight": "model-00031-of-00086.safetensors", + "model.layers.45.self_attn.v_proj.weight_scale": "model-00031-of-00086.safetensors", + "model.layers.46.input_layernorm.weight": "model-00033-of-00086.safetensors", + "model.layers.46.mlp.down_proj.weight": "model-00033-of-00086.safetensors", + "model.layers.46.mlp.down_proj.weight_scale": "model-00033-of-00086.safetensors", + "model.layers.46.mlp.gate_proj.weight": "model-00032-of-00086.safetensors", + "model.layers.46.mlp.gate_proj.weight_scale": "model-00032-of-00086.safetensors", + "model.layers.46.mlp.up_proj.weight": "model-00032-of-00086.safetensors", + "model.layers.46.mlp.up_proj.weight_scale": "model-00032-of-00086.safetensors", + "model.layers.46.post_attention_layernorm.weight": "model-00033-of-00086.safetensors", + "model.layers.46.self_attn.k_proj.weight": "model-00032-of-00086.safetensors", + "model.layers.46.self_attn.k_proj.weight_scale": "model-00032-of-00086.safetensors", + "model.layers.46.self_attn.o_proj.weight": "model-00032-of-00086.safetensors", + "model.layers.46.self_attn.o_proj.weight_scale": "model-00032-of-00086.safetensors", + "model.layers.46.self_attn.q_proj.weight": "model-00032-of-00086.safetensors", + "model.layers.46.self_attn.q_proj.weight_scale": "model-00032-of-00086.safetensors", + "model.layers.46.self_attn.v_proj.weight": "model-00032-of-00086.safetensors", + "model.layers.46.self_attn.v_proj.weight_scale": "model-00032-of-00086.safetensors", + "model.layers.47.input_layernorm.weight": "model-00033-of-00086.safetensors", + "model.layers.47.mlp.down_proj.weight": "model-00033-of-00086.safetensors", + "model.layers.47.mlp.down_proj.weight_scale": "model-00033-of-00086.safetensors", + "model.layers.47.mlp.gate_proj.weight": "model-00033-of-00086.safetensors", + "model.layers.47.mlp.gate_proj.weight_scale": "model-00033-of-00086.safetensors", + "model.layers.47.mlp.up_proj.weight": "model-00033-of-00086.safetensors", + "model.layers.47.mlp.up_proj.weight_scale": "model-00033-of-00086.safetensors", + "model.layers.47.post_attention_layernorm.weight": "model-00033-of-00086.safetensors", + "model.layers.47.self_attn.k_proj.weight": "model-00033-of-00086.safetensors", + "model.layers.47.self_attn.k_proj.weight_scale": "model-00033-of-00086.safetensors", + "model.layers.47.self_attn.o_proj.weight": "model-00033-of-00086.safetensors", + "model.layers.47.self_attn.o_proj.weight_scale": "model-00033-of-00086.safetensors", + "model.layers.47.self_attn.q_proj.weight": "model-00033-of-00086.safetensors", + "model.layers.47.self_attn.q_proj.weight_scale": "model-00033-of-00086.safetensors", + "model.layers.47.self_attn.v_proj.weight": "model-00033-of-00086.safetensors", + "model.layers.47.self_attn.v_proj.weight_scale": "model-00033-of-00086.safetensors", + "model.layers.48.input_layernorm.weight": "model-00034-of-00086.safetensors", + "model.layers.48.mlp.down_proj.weight": "model-00034-of-00086.safetensors", + "model.layers.48.mlp.down_proj.weight_scale": "model-00034-of-00086.safetensors", + "model.layers.48.mlp.gate_proj.weight": "model-00034-of-00086.safetensors", + "model.layers.48.mlp.gate_proj.weight_scale": "model-00034-of-00086.safetensors", + "model.layers.48.mlp.up_proj.weight": "model-00034-of-00086.safetensors", + "model.layers.48.mlp.up_proj.weight_scale": "model-00034-of-00086.safetensors", + "model.layers.48.post_attention_layernorm.weight": "model-00034-of-00086.safetensors", + "model.layers.48.self_attn.k_proj.weight": "model-00033-of-00086.safetensors", + "model.layers.48.self_attn.k_proj.weight_scale": "model-00033-of-00086.safetensors", + "model.layers.48.self_attn.o_proj.weight": "model-00033-of-00086.safetensors", + "model.layers.48.self_attn.o_proj.weight_scale": "model-00033-of-00086.safetensors", + "model.layers.48.self_attn.q_proj.weight": "model-00033-of-00086.safetensors", + "model.layers.48.self_attn.q_proj.weight_scale": "model-00033-of-00086.safetensors", + "model.layers.48.self_attn.v_proj.weight": "model-00033-of-00086.safetensors", + "model.layers.48.self_attn.v_proj.weight_scale": "model-00033-of-00086.safetensors", + "model.layers.49.input_layernorm.weight": "model-00035-of-00086.safetensors", + "model.layers.49.mlp.down_proj.weight": "model-00035-of-00086.safetensors", + "model.layers.49.mlp.down_proj.weight_scale": "model-00035-of-00086.safetensors", + "model.layers.49.mlp.gate_proj.weight": "model-00034-of-00086.safetensors", + "model.layers.49.mlp.gate_proj.weight_scale": "model-00034-of-00086.safetensors", + "model.layers.49.mlp.up_proj.weight": "model-00034-of-00086.safetensors", + "model.layers.49.mlp.up_proj.weight_scale": "model-00034-of-00086.safetensors", + "model.layers.49.post_attention_layernorm.weight": "model-00035-of-00086.safetensors", + "model.layers.49.self_attn.k_proj.weight": "model-00034-of-00086.safetensors", + "model.layers.49.self_attn.k_proj.weight_scale": "model-00034-of-00086.safetensors", + "model.layers.49.self_attn.o_proj.weight": "model-00034-of-00086.safetensors", + "model.layers.49.self_attn.o_proj.weight_scale": "model-00034-of-00086.safetensors", + "model.layers.49.self_attn.q_proj.weight": "model-00034-of-00086.safetensors", + "model.layers.49.self_attn.q_proj.weight_scale": "model-00034-of-00086.safetensors", + "model.layers.49.self_attn.v_proj.weight": "model-00034-of-00086.safetensors", + "model.layers.49.self_attn.v_proj.weight_scale": "model-00034-of-00086.safetensors", + "model.layers.5.input_layernorm.weight": "model-00005-of-00086.safetensors", + "model.layers.5.mlp.down_proj.weight": "model-00005-of-00086.safetensors", + "model.layers.5.mlp.down_proj.weight_scale": "model-00005-of-00086.safetensors", + "model.layers.5.mlp.gate_proj.weight": "model-00005-of-00086.safetensors", + "model.layers.5.mlp.gate_proj.weight_scale": "model-00005-of-00086.safetensors", + "model.layers.5.mlp.up_proj.weight": "model-00005-of-00086.safetensors", + "model.layers.5.mlp.up_proj.weight_scale": "model-00005-of-00086.safetensors", + "model.layers.5.post_attention_layernorm.weight": "model-00005-of-00086.safetensors", + "model.layers.5.self_attn.k_proj.weight": "model-00005-of-00086.safetensors", + "model.layers.5.self_attn.k_proj.weight_scale": "model-00005-of-00086.safetensors", + "model.layers.5.self_attn.o_proj.weight": "model-00005-of-00086.safetensors", + "model.layers.5.self_attn.o_proj.weight_scale": "model-00005-of-00086.safetensors", + "model.layers.5.self_attn.q_proj.weight": "model-00005-of-00086.safetensors", + "model.layers.5.self_attn.q_proj.weight_scale": "model-00005-of-00086.safetensors", + "model.layers.5.self_attn.v_proj.weight": "model-00005-of-00086.safetensors", + "model.layers.5.self_attn.v_proj.weight_scale": "model-00005-of-00086.safetensors", + "model.layers.50.input_layernorm.weight": "model-00035-of-00086.safetensors", + "model.layers.50.mlp.down_proj.weight": "model-00035-of-00086.safetensors", + "model.layers.50.mlp.down_proj.weight_scale": "model-00035-of-00086.safetensors", + "model.layers.50.mlp.gate_proj.weight": "model-00035-of-00086.safetensors", + "model.layers.50.mlp.gate_proj.weight_scale": "model-00035-of-00086.safetensors", + "model.layers.50.mlp.up_proj.weight": "model-00035-of-00086.safetensors", + "model.layers.50.mlp.up_proj.weight_scale": "model-00035-of-00086.safetensors", + "model.layers.50.post_attention_layernorm.weight": "model-00035-of-00086.safetensors", + "model.layers.50.self_attn.k_proj.weight": "model-00035-of-00086.safetensors", + "model.layers.50.self_attn.k_proj.weight_scale": "model-00035-of-00086.safetensors", + "model.layers.50.self_attn.o_proj.weight": "model-00035-of-00086.safetensors", + "model.layers.50.self_attn.o_proj.weight_scale": "model-00035-of-00086.safetensors", + "model.layers.50.self_attn.q_proj.weight": "model-00035-of-00086.safetensors", + "model.layers.50.self_attn.q_proj.weight_scale": "model-00035-of-00086.safetensors", + "model.layers.50.self_attn.v_proj.weight": "model-00035-of-00086.safetensors", + "model.layers.50.self_attn.v_proj.weight_scale": "model-00035-of-00086.safetensors", + "model.layers.51.input_layernorm.weight": "model-00036-of-00086.safetensors", + "model.layers.51.mlp.down_proj.weight": "model-00036-of-00086.safetensors", + "model.layers.51.mlp.down_proj.weight_scale": "model-00036-of-00086.safetensors", + "model.layers.51.mlp.gate_proj.weight": "model-00036-of-00086.safetensors", + "model.layers.51.mlp.gate_proj.weight_scale": "model-00036-of-00086.safetensors", + "model.layers.51.mlp.up_proj.weight": "model-00036-of-00086.safetensors", + "model.layers.51.mlp.up_proj.weight_scale": "model-00036-of-00086.safetensors", + "model.layers.51.post_attention_layernorm.weight": "model-00036-of-00086.safetensors", + "model.layers.51.self_attn.k_proj.weight": "model-00035-of-00086.safetensors", + "model.layers.51.self_attn.k_proj.weight_scale": "model-00035-of-00086.safetensors", + "model.layers.51.self_attn.o_proj.weight": "model-00035-of-00086.safetensors", + "model.layers.51.self_attn.o_proj.weight_scale": "model-00035-of-00086.safetensors", + "model.layers.51.self_attn.q_proj.weight": "model-00035-of-00086.safetensors", + "model.layers.51.self_attn.q_proj.weight_scale": "model-00035-of-00086.safetensors", + "model.layers.51.self_attn.v_proj.weight": "model-00035-of-00086.safetensors", + "model.layers.51.self_attn.v_proj.weight_scale": "model-00035-of-00086.safetensors", + "model.layers.52.input_layernorm.weight": "model-00037-of-00086.safetensors", + "model.layers.52.mlp.down_proj.weight": "model-00037-of-00086.safetensors", + "model.layers.52.mlp.down_proj.weight_scale": "model-00037-of-00086.safetensors", + "model.layers.52.mlp.gate_proj.weight": "model-00036-of-00086.safetensors", + "model.layers.52.mlp.gate_proj.weight_scale": "model-00036-of-00086.safetensors", + "model.layers.52.mlp.up_proj.weight": "model-00036-of-00086.safetensors", + "model.layers.52.mlp.up_proj.weight_scale": "model-00036-of-00086.safetensors", + "model.layers.52.post_attention_layernorm.weight": "model-00037-of-00086.safetensors", + "model.layers.52.self_attn.k_proj.weight": "model-00036-of-00086.safetensors", + "model.layers.52.self_attn.k_proj.weight_scale": "model-00036-of-00086.safetensors", + "model.layers.52.self_attn.o_proj.weight": "model-00036-of-00086.safetensors", + "model.layers.52.self_attn.o_proj.weight_scale": "model-00036-of-00086.safetensors", + "model.layers.52.self_attn.q_proj.weight": "model-00036-of-00086.safetensors", + "model.layers.52.self_attn.q_proj.weight_scale": "model-00036-of-00086.safetensors", + "model.layers.52.self_attn.v_proj.weight": "model-00036-of-00086.safetensors", + "model.layers.52.self_attn.v_proj.weight_scale": "model-00036-of-00086.safetensors", + "model.layers.53.input_layernorm.weight": "model-00037-of-00086.safetensors", + "model.layers.53.mlp.down_proj.weight": "model-00037-of-00086.safetensors", + "model.layers.53.mlp.down_proj.weight_scale": "model-00037-of-00086.safetensors", + "model.layers.53.mlp.gate_proj.weight": "model-00037-of-00086.safetensors", + "model.layers.53.mlp.gate_proj.weight_scale": "model-00037-of-00086.safetensors", + "model.layers.53.mlp.up_proj.weight": "model-00037-of-00086.safetensors", + "model.layers.53.mlp.up_proj.weight_scale": "model-00037-of-00086.safetensors", + "model.layers.53.post_attention_layernorm.weight": "model-00037-of-00086.safetensors", + "model.layers.53.self_attn.k_proj.weight": "model-00037-of-00086.safetensors", + "model.layers.53.self_attn.k_proj.weight_scale": "model-00037-of-00086.safetensors", + "model.layers.53.self_attn.o_proj.weight": "model-00037-of-00086.safetensors", + "model.layers.53.self_attn.o_proj.weight_scale": "model-00037-of-00086.safetensors", + "model.layers.53.self_attn.q_proj.weight": "model-00037-of-00086.safetensors", + "model.layers.53.self_attn.q_proj.weight_scale": "model-00037-of-00086.safetensors", + "model.layers.53.self_attn.v_proj.weight": "model-00037-of-00086.safetensors", + "model.layers.53.self_attn.v_proj.weight_scale": "model-00037-of-00086.safetensors", + "model.layers.54.input_layernorm.weight": "model-00038-of-00086.safetensors", + "model.layers.54.mlp.down_proj.weight": "model-00038-of-00086.safetensors", + "model.layers.54.mlp.down_proj.weight_scale": "model-00038-of-00086.safetensors", + "model.layers.54.mlp.gate_proj.weight": "model-00038-of-00086.safetensors", + "model.layers.54.mlp.gate_proj.weight_scale": "model-00038-of-00086.safetensors", + "model.layers.54.mlp.up_proj.weight": "model-00038-of-00086.safetensors", + "model.layers.54.mlp.up_proj.weight_scale": "model-00038-of-00086.safetensors", + "model.layers.54.post_attention_layernorm.weight": "model-00038-of-00086.safetensors", + "model.layers.54.self_attn.k_proj.weight": "model-00037-of-00086.safetensors", + "model.layers.54.self_attn.k_proj.weight_scale": "model-00037-of-00086.safetensors", + "model.layers.54.self_attn.o_proj.weight": "model-00037-of-00086.safetensors", + "model.layers.54.self_attn.o_proj.weight_scale": "model-00037-of-00086.safetensors", + "model.layers.54.self_attn.q_proj.weight": "model-00037-of-00086.safetensors", + "model.layers.54.self_attn.q_proj.weight_scale": "model-00037-of-00086.safetensors", + "model.layers.54.self_attn.v_proj.weight": "model-00037-of-00086.safetensors", + "model.layers.54.self_attn.v_proj.weight_scale": "model-00037-of-00086.safetensors", + "model.layers.55.input_layernorm.weight": "model-00039-of-00086.safetensors", + "model.layers.55.mlp.down_proj.weight": "model-00039-of-00086.safetensors", + "model.layers.55.mlp.down_proj.weight_scale": "model-00039-of-00086.safetensors", + "model.layers.55.mlp.gate_proj.weight": "model-00038-of-00086.safetensors", + "model.layers.55.mlp.gate_proj.weight_scale": "model-00038-of-00086.safetensors", + "model.layers.55.mlp.up_proj.weight": "model-00038-of-00086.safetensors", + "model.layers.55.mlp.up_proj.weight_scale": "model-00038-of-00086.safetensors", + "model.layers.55.post_attention_layernorm.weight": "model-00039-of-00086.safetensors", + "model.layers.55.self_attn.k_proj.weight": "model-00038-of-00086.safetensors", + "model.layers.55.self_attn.k_proj.weight_scale": "model-00038-of-00086.safetensors", + "model.layers.55.self_attn.o_proj.weight": "model-00038-of-00086.safetensors", + "model.layers.55.self_attn.o_proj.weight_scale": "model-00038-of-00086.safetensors", + "model.layers.55.self_attn.q_proj.weight": "model-00038-of-00086.safetensors", + "model.layers.55.self_attn.q_proj.weight_scale": "model-00038-of-00086.safetensors", + "model.layers.55.self_attn.v_proj.weight": "model-00038-of-00086.safetensors", + "model.layers.55.self_attn.v_proj.weight_scale": "model-00038-of-00086.safetensors", + "model.layers.56.input_layernorm.weight": "model-00039-of-00086.safetensors", + "model.layers.56.mlp.down_proj.weight": "model-00039-of-00086.safetensors", + "model.layers.56.mlp.down_proj.weight_scale": "model-00039-of-00086.safetensors", + "model.layers.56.mlp.gate_proj.weight": "model-00039-of-00086.safetensors", + "model.layers.56.mlp.gate_proj.weight_scale": "model-00039-of-00086.safetensors", + "model.layers.56.mlp.up_proj.weight": "model-00039-of-00086.safetensors", + "model.layers.56.mlp.up_proj.weight_scale": "model-00039-of-00086.safetensors", + "model.layers.56.post_attention_layernorm.weight": "model-00039-of-00086.safetensors", + "model.layers.56.self_attn.k_proj.weight": "model-00039-of-00086.safetensors", + "model.layers.56.self_attn.k_proj.weight_scale": "model-00039-of-00086.safetensors", + "model.layers.56.self_attn.o_proj.weight": "model-00039-of-00086.safetensors", + "model.layers.56.self_attn.o_proj.weight_scale": "model-00039-of-00086.safetensors", + "model.layers.56.self_attn.q_proj.weight": "model-00039-of-00086.safetensors", + "model.layers.56.self_attn.q_proj.weight_scale": "model-00039-of-00086.safetensors", + "model.layers.56.self_attn.v_proj.weight": "model-00039-of-00086.safetensors", + "model.layers.56.self_attn.v_proj.weight_scale": "model-00039-of-00086.safetensors", + "model.layers.57.input_layernorm.weight": "model-00040-of-00086.safetensors", + "model.layers.57.mlp.down_proj.weight": "model-00040-of-00086.safetensors", + "model.layers.57.mlp.down_proj.weight_scale": "model-00040-of-00086.safetensors", + "model.layers.57.mlp.gate_proj.weight": "model-00040-of-00086.safetensors", + "model.layers.57.mlp.gate_proj.weight_scale": "model-00040-of-00086.safetensors", + "model.layers.57.mlp.up_proj.weight": "model-00040-of-00086.safetensors", + "model.layers.57.mlp.up_proj.weight_scale": "model-00040-of-00086.safetensors", + "model.layers.57.post_attention_layernorm.weight": "model-00040-of-00086.safetensors", + "model.layers.57.self_attn.k_proj.weight": "model-00039-of-00086.safetensors", + "model.layers.57.self_attn.k_proj.weight_scale": "model-00039-of-00086.safetensors", + "model.layers.57.self_attn.o_proj.weight": "model-00039-of-00086.safetensors", + "model.layers.57.self_attn.o_proj.weight_scale": "model-00039-of-00086.safetensors", + "model.layers.57.self_attn.q_proj.weight": "model-00039-of-00086.safetensors", + "model.layers.57.self_attn.q_proj.weight_scale": "model-00039-of-00086.safetensors", + "model.layers.57.self_attn.v_proj.weight": "model-00039-of-00086.safetensors", + "model.layers.57.self_attn.v_proj.weight_scale": "model-00039-of-00086.safetensors", + "model.layers.58.input_layernorm.weight": "model-00041-of-00086.safetensors", + "model.layers.58.mlp.down_proj.weight": "model-00041-of-00086.safetensors", + "model.layers.58.mlp.down_proj.weight_scale": "model-00041-of-00086.safetensors", + "model.layers.58.mlp.gate_proj.weight": "model-00040-of-00086.safetensors", + "model.layers.58.mlp.gate_proj.weight_scale": "model-00040-of-00086.safetensors", + "model.layers.58.mlp.up_proj.weight": "model-00040-of-00086.safetensors", + "model.layers.58.mlp.up_proj.weight_scale": "model-00040-of-00086.safetensors", + "model.layers.58.post_attention_layernorm.weight": "model-00041-of-00086.safetensors", + "model.layers.58.self_attn.k_proj.weight": "model-00040-of-00086.safetensors", + "model.layers.58.self_attn.k_proj.weight_scale": "model-00040-of-00086.safetensors", + "model.layers.58.self_attn.o_proj.weight": "model-00040-of-00086.safetensors", + "model.layers.58.self_attn.o_proj.weight_scale": "model-00040-of-00086.safetensors", + "model.layers.58.self_attn.q_proj.weight": "model-00040-of-00086.safetensors", + "model.layers.58.self_attn.q_proj.weight_scale": "model-00040-of-00086.safetensors", + "model.layers.58.self_attn.v_proj.weight": "model-00040-of-00086.safetensors", + "model.layers.58.self_attn.v_proj.weight_scale": "model-00040-of-00086.safetensors", + "model.layers.59.input_layernorm.weight": "model-00041-of-00086.safetensors", + "model.layers.59.mlp.down_proj.weight": "model-00041-of-00086.safetensors", + "model.layers.59.mlp.down_proj.weight_scale": "model-00041-of-00086.safetensors", + "model.layers.59.mlp.gate_proj.weight": "model-00041-of-00086.safetensors", + "model.layers.59.mlp.gate_proj.weight_scale": "model-00041-of-00086.safetensors", + "model.layers.59.mlp.up_proj.weight": "model-00041-of-00086.safetensors", + "model.layers.59.mlp.up_proj.weight_scale": "model-00041-of-00086.safetensors", + "model.layers.59.post_attention_layernorm.weight": "model-00041-of-00086.safetensors", + "model.layers.59.self_attn.k_proj.weight": "model-00041-of-00086.safetensors", + "model.layers.59.self_attn.k_proj.weight_scale": "model-00041-of-00086.safetensors", + "model.layers.59.self_attn.o_proj.weight": "model-00041-of-00086.safetensors", + "model.layers.59.self_attn.o_proj.weight_scale": "model-00041-of-00086.safetensors", + "model.layers.59.self_attn.q_proj.weight": "model-00041-of-00086.safetensors", + "model.layers.59.self_attn.q_proj.weight_scale": "model-00041-of-00086.safetensors", + "model.layers.59.self_attn.v_proj.weight": "model-00041-of-00086.safetensors", + "model.layers.59.self_attn.v_proj.weight_scale": "model-00041-of-00086.safetensors", + "model.layers.6.input_layernorm.weight": "model-00006-of-00086.safetensors", + "model.layers.6.mlp.down_proj.weight": "model-00006-of-00086.safetensors", + "model.layers.6.mlp.down_proj.weight_scale": "model-00006-of-00086.safetensors", + "model.layers.6.mlp.gate_proj.weight": "model-00006-of-00086.safetensors", + "model.layers.6.mlp.gate_proj.weight_scale": "model-00006-of-00086.safetensors", + "model.layers.6.mlp.up_proj.weight": "model-00006-of-00086.safetensors", + "model.layers.6.mlp.up_proj.weight_scale": "model-00006-of-00086.safetensors", + "model.layers.6.post_attention_layernorm.weight": "model-00006-of-00086.safetensors", + "model.layers.6.self_attn.k_proj.weight": "model-00005-of-00086.safetensors", + "model.layers.6.self_attn.k_proj.weight_scale": "model-00005-of-00086.safetensors", + "model.layers.6.self_attn.o_proj.weight": "model-00005-of-00086.safetensors", + "model.layers.6.self_attn.o_proj.weight_scale": "model-00005-of-00086.safetensors", + "model.layers.6.self_attn.q_proj.weight": "model-00005-of-00086.safetensors", + "model.layers.6.self_attn.q_proj.weight_scale": "model-00005-of-00086.safetensors", + "model.layers.6.self_attn.v_proj.weight": "model-00005-of-00086.safetensors", + "model.layers.6.self_attn.v_proj.weight_scale": "model-00005-of-00086.safetensors", + "model.layers.60.input_layernorm.weight": "model-00042-of-00086.safetensors", + "model.layers.60.mlp.down_proj.weight": "model-00042-of-00086.safetensors", + "model.layers.60.mlp.down_proj.weight_scale": "model-00042-of-00086.safetensors", + "model.layers.60.mlp.gate_proj.weight": "model-00042-of-00086.safetensors", + "model.layers.60.mlp.gate_proj.weight_scale": "model-00042-of-00086.safetensors", + "model.layers.60.mlp.up_proj.weight": "model-00042-of-00086.safetensors", + "model.layers.60.mlp.up_proj.weight_scale": "model-00042-of-00086.safetensors", + "model.layers.60.post_attention_layernorm.weight": "model-00042-of-00086.safetensors", + "model.layers.60.self_attn.k_proj.weight": "model-00041-of-00086.safetensors", + "model.layers.60.self_attn.k_proj.weight_scale": "model-00041-of-00086.safetensors", + "model.layers.60.self_attn.o_proj.weight": "model-00041-of-00086.safetensors", + "model.layers.60.self_attn.o_proj.weight_scale": "model-00041-of-00086.safetensors", + "model.layers.60.self_attn.q_proj.weight": "model-00041-of-00086.safetensors", + "model.layers.60.self_attn.q_proj.weight_scale": "model-00041-of-00086.safetensors", + "model.layers.60.self_attn.v_proj.weight": "model-00041-of-00086.safetensors", + "model.layers.60.self_attn.v_proj.weight_scale": "model-00041-of-00086.safetensors", + "model.layers.61.input_layernorm.weight": "model-00043-of-00086.safetensors", + "model.layers.61.mlp.down_proj.weight": "model-00043-of-00086.safetensors", + "model.layers.61.mlp.down_proj.weight_scale": "model-00043-of-00086.safetensors", + "model.layers.61.mlp.gate_proj.weight": "model-00042-of-00086.safetensors", + "model.layers.61.mlp.gate_proj.weight_scale": "model-00042-of-00086.safetensors", + "model.layers.61.mlp.up_proj.weight": "model-00042-of-00086.safetensors", + "model.layers.61.mlp.up_proj.weight_scale": "model-00042-of-00086.safetensors", + "model.layers.61.post_attention_layernorm.weight": "model-00043-of-00086.safetensors", + "model.layers.61.self_attn.k_proj.weight": "model-00042-of-00086.safetensors", + "model.layers.61.self_attn.k_proj.weight_scale": "model-00042-of-00086.safetensors", + "model.layers.61.self_attn.o_proj.weight": "model-00042-of-00086.safetensors", + "model.layers.61.self_attn.o_proj.weight_scale": "model-00042-of-00086.safetensors", + "model.layers.61.self_attn.q_proj.weight": "model-00042-of-00086.safetensors", + "model.layers.61.self_attn.q_proj.weight_scale": "model-00042-of-00086.safetensors", + "model.layers.61.self_attn.v_proj.weight": "model-00042-of-00086.safetensors", + "model.layers.61.self_attn.v_proj.weight_scale": "model-00042-of-00086.safetensors", + "model.layers.62.input_layernorm.weight": "model-00043-of-00086.safetensors", + "model.layers.62.mlp.down_proj.weight": "model-00043-of-00086.safetensors", + "model.layers.62.mlp.down_proj.weight_scale": "model-00043-of-00086.safetensors", + "model.layers.62.mlp.gate_proj.weight": "model-00043-of-00086.safetensors", + "model.layers.62.mlp.gate_proj.weight_scale": "model-00043-of-00086.safetensors", + "model.layers.62.mlp.up_proj.weight": "model-00043-of-00086.safetensors", + "model.layers.62.mlp.up_proj.weight_scale": "model-00043-of-00086.safetensors", + "model.layers.62.post_attention_layernorm.weight": "model-00043-of-00086.safetensors", + "model.layers.62.self_attn.k_proj.weight": "model-00043-of-00086.safetensors", + "model.layers.62.self_attn.k_proj.weight_scale": "model-00043-of-00086.safetensors", + "model.layers.62.self_attn.o_proj.weight": "model-00043-of-00086.safetensors", + "model.layers.62.self_attn.o_proj.weight_scale": "model-00043-of-00086.safetensors", + "model.layers.62.self_attn.q_proj.weight": "model-00043-of-00086.safetensors", + "model.layers.62.self_attn.q_proj.weight_scale": "model-00043-of-00086.safetensors", + "model.layers.62.self_attn.v_proj.weight": "model-00043-of-00086.safetensors", + "model.layers.62.self_attn.v_proj.weight_scale": "model-00043-of-00086.safetensors", + "model.layers.63.input_layernorm.weight": "model-00044-of-00086.safetensors", + "model.layers.63.mlp.down_proj.weight": "model-00044-of-00086.safetensors", + "model.layers.63.mlp.down_proj.weight_scale": "model-00044-of-00086.safetensors", + "model.layers.63.mlp.gate_proj.weight": "model-00044-of-00086.safetensors", + "model.layers.63.mlp.gate_proj.weight_scale": "model-00044-of-00086.safetensors", + "model.layers.63.mlp.up_proj.weight": "model-00044-of-00086.safetensors", + "model.layers.63.mlp.up_proj.weight_scale": "model-00044-of-00086.safetensors", + "model.layers.63.post_attention_layernorm.weight": "model-00044-of-00086.safetensors", + "model.layers.63.self_attn.k_proj.weight": "model-00043-of-00086.safetensors", + "model.layers.63.self_attn.k_proj.weight_scale": "model-00043-of-00086.safetensors", + "model.layers.63.self_attn.o_proj.weight": "model-00043-of-00086.safetensors", + "model.layers.63.self_attn.o_proj.weight_scale": "model-00043-of-00086.safetensors", + "model.layers.63.self_attn.q_proj.weight": "model-00043-of-00086.safetensors", + "model.layers.63.self_attn.q_proj.weight_scale": "model-00043-of-00086.safetensors", + "model.layers.63.self_attn.v_proj.weight": "model-00043-of-00086.safetensors", + "model.layers.63.self_attn.v_proj.weight_scale": "model-00043-of-00086.safetensors", + "model.layers.64.input_layernorm.weight": "model-00045-of-00086.safetensors", + "model.layers.64.mlp.down_proj.weight": "model-00045-of-00086.safetensors", + "model.layers.64.mlp.down_proj.weight_scale": "model-00045-of-00086.safetensors", + "model.layers.64.mlp.gate_proj.weight": "model-00044-of-00086.safetensors", + "model.layers.64.mlp.gate_proj.weight_scale": "model-00044-of-00086.safetensors", + "model.layers.64.mlp.up_proj.weight": "model-00044-of-00086.safetensors", + "model.layers.64.mlp.up_proj.weight_scale": "model-00044-of-00086.safetensors", + "model.layers.64.post_attention_layernorm.weight": "model-00045-of-00086.safetensors", + "model.layers.64.self_attn.k_proj.weight": "model-00044-of-00086.safetensors", + "model.layers.64.self_attn.k_proj.weight_scale": "model-00044-of-00086.safetensors", + "model.layers.64.self_attn.o_proj.weight": "model-00044-of-00086.safetensors", + "model.layers.64.self_attn.o_proj.weight_scale": "model-00044-of-00086.safetensors", + "model.layers.64.self_attn.q_proj.weight": "model-00044-of-00086.safetensors", + "model.layers.64.self_attn.q_proj.weight_scale": "model-00044-of-00086.safetensors", + "model.layers.64.self_attn.v_proj.weight": "model-00044-of-00086.safetensors", + "model.layers.64.self_attn.v_proj.weight_scale": "model-00044-of-00086.safetensors", + "model.layers.65.input_layernorm.weight": "model-00045-of-00086.safetensors", + "model.layers.65.mlp.down_proj.weight": "model-00045-of-00086.safetensors", + "model.layers.65.mlp.down_proj.weight_scale": "model-00045-of-00086.safetensors", + "model.layers.65.mlp.gate_proj.weight": "model-00045-of-00086.safetensors", + "model.layers.65.mlp.gate_proj.weight_scale": "model-00045-of-00086.safetensors", + "model.layers.65.mlp.up_proj.weight": "model-00045-of-00086.safetensors", + "model.layers.65.mlp.up_proj.weight_scale": "model-00045-of-00086.safetensors", + "model.layers.65.post_attention_layernorm.weight": "model-00045-of-00086.safetensors", + "model.layers.65.self_attn.k_proj.weight": "model-00045-of-00086.safetensors", + "model.layers.65.self_attn.k_proj.weight_scale": "model-00045-of-00086.safetensors", + "model.layers.65.self_attn.o_proj.weight": "model-00045-of-00086.safetensors", + "model.layers.65.self_attn.o_proj.weight_scale": "model-00045-of-00086.safetensors", + "model.layers.65.self_attn.q_proj.weight": "model-00045-of-00086.safetensors", + "model.layers.65.self_attn.q_proj.weight_scale": "model-00045-of-00086.safetensors", + "model.layers.65.self_attn.v_proj.weight": "model-00045-of-00086.safetensors", + "model.layers.65.self_attn.v_proj.weight_scale": "model-00045-of-00086.safetensors", + "model.layers.66.input_layernorm.weight": "model-00046-of-00086.safetensors", + "model.layers.66.mlp.down_proj.weight": "model-00046-of-00086.safetensors", + "model.layers.66.mlp.down_proj.weight_scale": "model-00046-of-00086.safetensors", + "model.layers.66.mlp.gate_proj.weight": "model-00046-of-00086.safetensors", + "model.layers.66.mlp.gate_proj.weight_scale": "model-00046-of-00086.safetensors", + "model.layers.66.mlp.up_proj.weight": "model-00046-of-00086.safetensors", + "model.layers.66.mlp.up_proj.weight_scale": "model-00046-of-00086.safetensors", + "model.layers.66.post_attention_layernorm.weight": "model-00046-of-00086.safetensors", + "model.layers.66.self_attn.k_proj.weight": "model-00045-of-00086.safetensors", + "model.layers.66.self_attn.k_proj.weight_scale": "model-00045-of-00086.safetensors", + "model.layers.66.self_attn.o_proj.weight": "model-00045-of-00086.safetensors", + "model.layers.66.self_attn.o_proj.weight_scale": "model-00045-of-00086.safetensors", + "model.layers.66.self_attn.q_proj.weight": "model-00045-of-00086.safetensors", + "model.layers.66.self_attn.q_proj.weight_scale": "model-00045-of-00086.safetensors", + "model.layers.66.self_attn.v_proj.weight": "model-00045-of-00086.safetensors", + "model.layers.66.self_attn.v_proj.weight_scale": "model-00045-of-00086.safetensors", + "model.layers.67.input_layernorm.weight": "model-00047-of-00086.safetensors", + "model.layers.67.mlp.down_proj.weight": "model-00047-of-00086.safetensors", + "model.layers.67.mlp.down_proj.weight_scale": "model-00047-of-00086.safetensors", + "model.layers.67.mlp.gate_proj.weight": "model-00046-of-00086.safetensors", + "model.layers.67.mlp.gate_proj.weight_scale": "model-00046-of-00086.safetensors", + "model.layers.67.mlp.up_proj.weight": "model-00046-of-00086.safetensors", + "model.layers.67.mlp.up_proj.weight_scale": "model-00046-of-00086.safetensors", + "model.layers.67.post_attention_layernorm.weight": "model-00047-of-00086.safetensors", + "model.layers.67.self_attn.k_proj.weight": "model-00046-of-00086.safetensors", + "model.layers.67.self_attn.k_proj.weight_scale": "model-00046-of-00086.safetensors", + "model.layers.67.self_attn.o_proj.weight": "model-00046-of-00086.safetensors", + "model.layers.67.self_attn.o_proj.weight_scale": "model-00046-of-00086.safetensors", + "model.layers.67.self_attn.q_proj.weight": "model-00046-of-00086.safetensors", + "model.layers.67.self_attn.q_proj.weight_scale": "model-00046-of-00086.safetensors", + "model.layers.67.self_attn.v_proj.weight": "model-00046-of-00086.safetensors", + "model.layers.67.self_attn.v_proj.weight_scale": "model-00046-of-00086.safetensors", + "model.layers.68.input_layernorm.weight": "model-00047-of-00086.safetensors", + "model.layers.68.mlp.down_proj.weight": "model-00047-of-00086.safetensors", + "model.layers.68.mlp.down_proj.weight_scale": "model-00047-of-00086.safetensors", + "model.layers.68.mlp.gate_proj.weight": "model-00047-of-00086.safetensors", + "model.layers.68.mlp.gate_proj.weight_scale": "model-00047-of-00086.safetensors", + "model.layers.68.mlp.up_proj.weight": "model-00047-of-00086.safetensors", + "model.layers.68.mlp.up_proj.weight_scale": "model-00047-of-00086.safetensors", + "model.layers.68.post_attention_layernorm.weight": "model-00047-of-00086.safetensors", + "model.layers.68.self_attn.k_proj.weight": "model-00047-of-00086.safetensors", + "model.layers.68.self_attn.k_proj.weight_scale": "model-00047-of-00086.safetensors", + "model.layers.68.self_attn.o_proj.weight": "model-00047-of-00086.safetensors", + "model.layers.68.self_attn.o_proj.weight_scale": "model-00047-of-00086.safetensors", + "model.layers.68.self_attn.q_proj.weight": "model-00047-of-00086.safetensors", + "model.layers.68.self_attn.q_proj.weight_scale": "model-00047-of-00086.safetensors", + "model.layers.68.self_attn.v_proj.weight": "model-00047-of-00086.safetensors", + "model.layers.68.self_attn.v_proj.weight_scale": "model-00047-of-00086.safetensors", + "model.layers.69.input_layernorm.weight": "model-00048-of-00086.safetensors", + "model.layers.69.mlp.down_proj.weight": "model-00048-of-00086.safetensors", + "model.layers.69.mlp.down_proj.weight_scale": "model-00048-of-00086.safetensors", + "model.layers.69.mlp.gate_proj.weight": "model-00048-of-00086.safetensors", + "model.layers.69.mlp.gate_proj.weight_scale": "model-00048-of-00086.safetensors", + "model.layers.69.mlp.up_proj.weight": "model-00048-of-00086.safetensors", + "model.layers.69.mlp.up_proj.weight_scale": "model-00048-of-00086.safetensors", + "model.layers.69.post_attention_layernorm.weight": "model-00048-of-00086.safetensors", + "model.layers.69.self_attn.k_proj.weight": "model-00047-of-00086.safetensors", + "model.layers.69.self_attn.k_proj.weight_scale": "model-00047-of-00086.safetensors", + "model.layers.69.self_attn.o_proj.weight": "model-00047-of-00086.safetensors", + "model.layers.69.self_attn.o_proj.weight_scale": "model-00047-of-00086.safetensors", + "model.layers.69.self_attn.q_proj.weight": "model-00047-of-00086.safetensors", + "model.layers.69.self_attn.q_proj.weight_scale": "model-00047-of-00086.safetensors", + "model.layers.69.self_attn.v_proj.weight": "model-00047-of-00086.safetensors", + "model.layers.69.self_attn.v_proj.weight_scale": "model-00047-of-00086.safetensors", + "model.layers.7.input_layernorm.weight": "model-00007-of-00086.safetensors", + "model.layers.7.mlp.down_proj.weight": "model-00007-of-00086.safetensors", + "model.layers.7.mlp.down_proj.weight_scale": "model-00007-of-00086.safetensors", + "model.layers.7.mlp.gate_proj.weight": "model-00006-of-00086.safetensors", + "model.layers.7.mlp.gate_proj.weight_scale": "model-00006-of-00086.safetensors", + "model.layers.7.mlp.up_proj.weight": "model-00006-of-00086.safetensors", + "model.layers.7.mlp.up_proj.weight_scale": "model-00006-of-00086.safetensors", + "model.layers.7.post_attention_layernorm.weight": "model-00007-of-00086.safetensors", + "model.layers.7.self_attn.k_proj.weight": "model-00006-of-00086.safetensors", + "model.layers.7.self_attn.k_proj.weight_scale": "model-00006-of-00086.safetensors", + "model.layers.7.self_attn.o_proj.weight": "model-00006-of-00086.safetensors", + "model.layers.7.self_attn.o_proj.weight_scale": "model-00006-of-00086.safetensors", + "model.layers.7.self_attn.q_proj.weight": "model-00006-of-00086.safetensors", + "model.layers.7.self_attn.q_proj.weight_scale": "model-00006-of-00086.safetensors", + "model.layers.7.self_attn.v_proj.weight": "model-00006-of-00086.safetensors", + "model.layers.7.self_attn.v_proj.weight_scale": "model-00006-of-00086.safetensors", + "model.layers.70.input_layernorm.weight": "model-00049-of-00086.safetensors", + "model.layers.70.mlp.down_proj.weight": "model-00049-of-00086.safetensors", + "model.layers.70.mlp.down_proj.weight_scale": "model-00049-of-00086.safetensors", + "model.layers.70.mlp.gate_proj.weight": "model-00048-of-00086.safetensors", + "model.layers.70.mlp.gate_proj.weight_scale": "model-00048-of-00086.safetensors", + "model.layers.70.mlp.up_proj.weight": "model-00048-of-00086.safetensors", + "model.layers.70.mlp.up_proj.weight_scale": "model-00048-of-00086.safetensors", + "model.layers.70.post_attention_layernorm.weight": "model-00049-of-00086.safetensors", + "model.layers.70.self_attn.k_proj.weight": "model-00048-of-00086.safetensors", + "model.layers.70.self_attn.k_proj.weight_scale": "model-00048-of-00086.safetensors", + "model.layers.70.self_attn.o_proj.weight": "model-00048-of-00086.safetensors", + "model.layers.70.self_attn.o_proj.weight_scale": "model-00048-of-00086.safetensors", + "model.layers.70.self_attn.q_proj.weight": "model-00048-of-00086.safetensors", + "model.layers.70.self_attn.q_proj.weight_scale": "model-00048-of-00086.safetensors", + "model.layers.70.self_attn.v_proj.weight": "model-00048-of-00086.safetensors", + "model.layers.70.self_attn.v_proj.weight_scale": "model-00048-of-00086.safetensors", + "model.layers.71.input_layernorm.weight": "model-00049-of-00086.safetensors", + "model.layers.71.mlp.down_proj.weight": "model-00049-of-00086.safetensors", + "model.layers.71.mlp.down_proj.weight_scale": "model-00049-of-00086.safetensors", + "model.layers.71.mlp.gate_proj.weight": "model-00049-of-00086.safetensors", + "model.layers.71.mlp.gate_proj.weight_scale": "model-00049-of-00086.safetensors", + "model.layers.71.mlp.up_proj.weight": "model-00049-of-00086.safetensors", + "model.layers.71.mlp.up_proj.weight_scale": "model-00049-of-00086.safetensors", + "model.layers.71.post_attention_layernorm.weight": "model-00049-of-00086.safetensors", + "model.layers.71.self_attn.k_proj.weight": "model-00049-of-00086.safetensors", + "model.layers.71.self_attn.k_proj.weight_scale": "model-00049-of-00086.safetensors", + "model.layers.71.self_attn.o_proj.weight": "model-00049-of-00086.safetensors", + "model.layers.71.self_attn.o_proj.weight_scale": "model-00049-of-00086.safetensors", + "model.layers.71.self_attn.q_proj.weight": "model-00049-of-00086.safetensors", + "model.layers.71.self_attn.q_proj.weight_scale": "model-00049-of-00086.safetensors", + "model.layers.71.self_attn.v_proj.weight": "model-00049-of-00086.safetensors", + "model.layers.71.self_attn.v_proj.weight_scale": "model-00049-of-00086.safetensors", + "model.layers.72.input_layernorm.weight": "model-00050-of-00086.safetensors", + "model.layers.72.mlp.down_proj.weight": "model-00050-of-00086.safetensors", + "model.layers.72.mlp.down_proj.weight_scale": "model-00050-of-00086.safetensors", + "model.layers.72.mlp.gate_proj.weight": "model-00050-of-00086.safetensors", + "model.layers.72.mlp.gate_proj.weight_scale": "model-00050-of-00086.safetensors", + "model.layers.72.mlp.up_proj.weight": "model-00050-of-00086.safetensors", + "model.layers.72.mlp.up_proj.weight_scale": "model-00050-of-00086.safetensors", + "model.layers.72.post_attention_layernorm.weight": "model-00050-of-00086.safetensors", + "model.layers.72.self_attn.k_proj.weight": "model-00049-of-00086.safetensors", + "model.layers.72.self_attn.k_proj.weight_scale": "model-00049-of-00086.safetensors", + "model.layers.72.self_attn.o_proj.weight": "model-00049-of-00086.safetensors", + "model.layers.72.self_attn.o_proj.weight_scale": "model-00049-of-00086.safetensors", + "model.layers.72.self_attn.q_proj.weight": "model-00049-of-00086.safetensors", + "model.layers.72.self_attn.q_proj.weight_scale": "model-00049-of-00086.safetensors", + "model.layers.72.self_attn.v_proj.weight": "model-00049-of-00086.safetensors", + "model.layers.72.self_attn.v_proj.weight_scale": "model-00049-of-00086.safetensors", + "model.layers.73.input_layernorm.weight": "model-00051-of-00086.safetensors", + "model.layers.73.mlp.down_proj.weight": "model-00051-of-00086.safetensors", + "model.layers.73.mlp.down_proj.weight_scale": "model-00051-of-00086.safetensors", + "model.layers.73.mlp.gate_proj.weight": "model-00050-of-00086.safetensors", + "model.layers.73.mlp.gate_proj.weight_scale": "model-00050-of-00086.safetensors", + "model.layers.73.mlp.up_proj.weight": "model-00050-of-00086.safetensors", + "model.layers.73.mlp.up_proj.weight_scale": "model-00050-of-00086.safetensors", + "model.layers.73.post_attention_layernorm.weight": "model-00051-of-00086.safetensors", + "model.layers.73.self_attn.k_proj.weight": "model-00050-of-00086.safetensors", + "model.layers.73.self_attn.k_proj.weight_scale": "model-00050-of-00086.safetensors", + "model.layers.73.self_attn.o_proj.weight": "model-00050-of-00086.safetensors", + "model.layers.73.self_attn.o_proj.weight_scale": "model-00050-of-00086.safetensors", + "model.layers.73.self_attn.q_proj.weight": "model-00050-of-00086.safetensors", + "model.layers.73.self_attn.q_proj.weight_scale": "model-00050-of-00086.safetensors", + "model.layers.73.self_attn.v_proj.weight": "model-00050-of-00086.safetensors", + "model.layers.73.self_attn.v_proj.weight_scale": "model-00050-of-00086.safetensors", + "model.layers.74.input_layernorm.weight": "model-00051-of-00086.safetensors", + "model.layers.74.mlp.down_proj.weight": "model-00051-of-00086.safetensors", + "model.layers.74.mlp.down_proj.weight_scale": "model-00051-of-00086.safetensors", + "model.layers.74.mlp.gate_proj.weight": "model-00051-of-00086.safetensors", + "model.layers.74.mlp.gate_proj.weight_scale": "model-00051-of-00086.safetensors", + "model.layers.74.mlp.up_proj.weight": "model-00051-of-00086.safetensors", + "model.layers.74.mlp.up_proj.weight_scale": "model-00051-of-00086.safetensors", + "model.layers.74.post_attention_layernorm.weight": "model-00051-of-00086.safetensors", + "model.layers.74.self_attn.k_proj.weight": "model-00051-of-00086.safetensors", + "model.layers.74.self_attn.k_proj.weight_scale": "model-00051-of-00086.safetensors", + "model.layers.74.self_attn.o_proj.weight": "model-00051-of-00086.safetensors", + "model.layers.74.self_attn.o_proj.weight_scale": "model-00051-of-00086.safetensors", + "model.layers.74.self_attn.q_proj.weight": "model-00051-of-00086.safetensors", + "model.layers.74.self_attn.q_proj.weight_scale": "model-00051-of-00086.safetensors", + "model.layers.74.self_attn.v_proj.weight": "model-00051-of-00086.safetensors", + "model.layers.74.self_attn.v_proj.weight_scale": "model-00051-of-00086.safetensors", + "model.layers.75.input_layernorm.weight": "model-00052-of-00086.safetensors", + "model.layers.75.mlp.down_proj.weight": "model-00052-of-00086.safetensors", + "model.layers.75.mlp.down_proj.weight_scale": "model-00052-of-00086.safetensors", + "model.layers.75.mlp.gate_proj.weight": "model-00052-of-00086.safetensors", + "model.layers.75.mlp.gate_proj.weight_scale": "model-00052-of-00086.safetensors", + "model.layers.75.mlp.up_proj.weight": "model-00052-of-00086.safetensors", + "model.layers.75.mlp.up_proj.weight_scale": "model-00052-of-00086.safetensors", + "model.layers.75.post_attention_layernorm.weight": "model-00052-of-00086.safetensors", + "model.layers.75.self_attn.k_proj.weight": "model-00051-of-00086.safetensors", + "model.layers.75.self_attn.k_proj.weight_scale": "model-00051-of-00086.safetensors", + "model.layers.75.self_attn.o_proj.weight": "model-00051-of-00086.safetensors", + "model.layers.75.self_attn.o_proj.weight_scale": "model-00051-of-00086.safetensors", + "model.layers.75.self_attn.q_proj.weight": "model-00051-of-00086.safetensors", + "model.layers.75.self_attn.q_proj.weight_scale": "model-00051-of-00086.safetensors", + "model.layers.75.self_attn.v_proj.weight": "model-00051-of-00086.safetensors", + "model.layers.75.self_attn.v_proj.weight_scale": "model-00051-of-00086.safetensors", + "model.layers.76.input_layernorm.weight": "model-00053-of-00086.safetensors", + "model.layers.76.mlp.down_proj.weight": "model-00053-of-00086.safetensors", + "model.layers.76.mlp.down_proj.weight_scale": "model-00053-of-00086.safetensors", + "model.layers.76.mlp.gate_proj.weight": "model-00052-of-00086.safetensors", + "model.layers.76.mlp.gate_proj.weight_scale": "model-00052-of-00086.safetensors", + "model.layers.76.mlp.up_proj.weight": "model-00052-of-00086.safetensors", + "model.layers.76.mlp.up_proj.weight_scale": "model-00052-of-00086.safetensors", + "model.layers.76.post_attention_layernorm.weight": "model-00053-of-00086.safetensors", + "model.layers.76.self_attn.k_proj.weight": "model-00052-of-00086.safetensors", + "model.layers.76.self_attn.k_proj.weight_scale": "model-00052-of-00086.safetensors", + "model.layers.76.self_attn.o_proj.weight": "model-00052-of-00086.safetensors", + "model.layers.76.self_attn.o_proj.weight_scale": "model-00052-of-00086.safetensors", + "model.layers.76.self_attn.q_proj.weight": "model-00052-of-00086.safetensors", + "model.layers.76.self_attn.q_proj.weight_scale": "model-00052-of-00086.safetensors", + "model.layers.76.self_attn.v_proj.weight": "model-00052-of-00086.safetensors", + "model.layers.76.self_attn.v_proj.weight_scale": "model-00052-of-00086.safetensors", + "model.layers.77.input_layernorm.weight": "model-00053-of-00086.safetensors", + "model.layers.77.mlp.down_proj.weight": "model-00053-of-00086.safetensors", + "model.layers.77.mlp.down_proj.weight_scale": "model-00053-of-00086.safetensors", + "model.layers.77.mlp.gate_proj.weight": "model-00053-of-00086.safetensors", + "model.layers.77.mlp.gate_proj.weight_scale": "model-00053-of-00086.safetensors", + "model.layers.77.mlp.up_proj.weight": "model-00053-of-00086.safetensors", + "model.layers.77.mlp.up_proj.weight_scale": "model-00053-of-00086.safetensors", + "model.layers.77.post_attention_layernorm.weight": "model-00053-of-00086.safetensors", + "model.layers.77.self_attn.k_proj.weight": "model-00053-of-00086.safetensors", + "model.layers.77.self_attn.k_proj.weight_scale": "model-00053-of-00086.safetensors", + "model.layers.77.self_attn.o_proj.weight": "model-00053-of-00086.safetensors", + "model.layers.77.self_attn.o_proj.weight_scale": "model-00053-of-00086.safetensors", + "model.layers.77.self_attn.q_proj.weight": "model-00053-of-00086.safetensors", + "model.layers.77.self_attn.q_proj.weight_scale": "model-00053-of-00086.safetensors", + "model.layers.77.self_attn.v_proj.weight": "model-00053-of-00086.safetensors", + "model.layers.77.self_attn.v_proj.weight_scale": "model-00053-of-00086.safetensors", + "model.layers.78.input_layernorm.weight": "model-00054-of-00086.safetensors", + "model.layers.78.mlp.down_proj.weight": "model-00054-of-00086.safetensors", + "model.layers.78.mlp.down_proj.weight_scale": "model-00054-of-00086.safetensors", + "model.layers.78.mlp.gate_proj.weight": "model-00054-of-00086.safetensors", + "model.layers.78.mlp.gate_proj.weight_scale": "model-00054-of-00086.safetensors", + "model.layers.78.mlp.up_proj.weight": "model-00054-of-00086.safetensors", + "model.layers.78.mlp.up_proj.weight_scale": "model-00054-of-00086.safetensors", + "model.layers.78.post_attention_layernorm.weight": "model-00054-of-00086.safetensors", + "model.layers.78.self_attn.k_proj.weight": "model-00053-of-00086.safetensors", + "model.layers.78.self_attn.k_proj.weight_scale": "model-00053-of-00086.safetensors", + "model.layers.78.self_attn.o_proj.weight": "model-00053-of-00086.safetensors", + "model.layers.78.self_attn.o_proj.weight_scale": "model-00053-of-00086.safetensors", + "model.layers.78.self_attn.q_proj.weight": "model-00053-of-00086.safetensors", + "model.layers.78.self_attn.q_proj.weight_scale": "model-00053-of-00086.safetensors", + "model.layers.78.self_attn.v_proj.weight": "model-00053-of-00086.safetensors", + "model.layers.78.self_attn.v_proj.weight_scale": "model-00053-of-00086.safetensors", + "model.layers.79.input_layernorm.weight": "model-00055-of-00086.safetensors", + "model.layers.79.mlp.down_proj.weight": "model-00055-of-00086.safetensors", + "model.layers.79.mlp.down_proj.weight_scale": "model-00055-of-00086.safetensors", + "model.layers.79.mlp.gate_proj.weight": "model-00054-of-00086.safetensors", + "model.layers.79.mlp.gate_proj.weight_scale": "model-00054-of-00086.safetensors", + "model.layers.79.mlp.up_proj.weight": "model-00054-of-00086.safetensors", + "model.layers.79.mlp.up_proj.weight_scale": "model-00054-of-00086.safetensors", + "model.layers.79.post_attention_layernorm.weight": "model-00055-of-00086.safetensors", + "model.layers.79.self_attn.k_proj.weight": "model-00054-of-00086.safetensors", + "model.layers.79.self_attn.k_proj.weight_scale": "model-00054-of-00086.safetensors", + "model.layers.79.self_attn.o_proj.weight": "model-00054-of-00086.safetensors", + "model.layers.79.self_attn.o_proj.weight_scale": "model-00054-of-00086.safetensors", + "model.layers.79.self_attn.q_proj.weight": "model-00054-of-00086.safetensors", + "model.layers.79.self_attn.q_proj.weight_scale": "model-00054-of-00086.safetensors", + "model.layers.79.self_attn.v_proj.weight": "model-00054-of-00086.safetensors", + "model.layers.79.self_attn.v_proj.weight_scale": "model-00054-of-00086.safetensors", + "model.layers.8.input_layernorm.weight": "model-00007-of-00086.safetensors", + "model.layers.8.mlp.down_proj.weight": "model-00007-of-00086.safetensors", + "model.layers.8.mlp.down_proj.weight_scale": "model-00007-of-00086.safetensors", + "model.layers.8.mlp.gate_proj.weight": "model-00007-of-00086.safetensors", + "model.layers.8.mlp.gate_proj.weight_scale": "model-00007-of-00086.safetensors", + "model.layers.8.mlp.up_proj.weight": "model-00007-of-00086.safetensors", + "model.layers.8.mlp.up_proj.weight_scale": "model-00007-of-00086.safetensors", + "model.layers.8.post_attention_layernorm.weight": "model-00007-of-00086.safetensors", + "model.layers.8.self_attn.k_proj.weight": "model-00007-of-00086.safetensors", + "model.layers.8.self_attn.k_proj.weight_scale": "model-00007-of-00086.safetensors", + "model.layers.8.self_attn.o_proj.weight": "model-00007-of-00086.safetensors", + "model.layers.8.self_attn.o_proj.weight_scale": "model-00007-of-00086.safetensors", + "model.layers.8.self_attn.q_proj.weight": "model-00007-of-00086.safetensors", + "model.layers.8.self_attn.q_proj.weight_scale": "model-00007-of-00086.safetensors", + "model.layers.8.self_attn.v_proj.weight": "model-00007-of-00086.safetensors", + "model.layers.8.self_attn.v_proj.weight_scale": "model-00007-of-00086.safetensors", + "model.layers.80.input_layernorm.weight": "model-00055-of-00086.safetensors", + "model.layers.80.mlp.down_proj.weight": "model-00055-of-00086.safetensors", + "model.layers.80.mlp.down_proj.weight_scale": "model-00055-of-00086.safetensors", + "model.layers.80.mlp.gate_proj.weight": "model-00055-of-00086.safetensors", + "model.layers.80.mlp.gate_proj.weight_scale": "model-00055-of-00086.safetensors", + "model.layers.80.mlp.up_proj.weight": "model-00055-of-00086.safetensors", + "model.layers.80.mlp.up_proj.weight_scale": "model-00055-of-00086.safetensors", + "model.layers.80.post_attention_layernorm.weight": "model-00055-of-00086.safetensors", + "model.layers.80.self_attn.k_proj.weight": "model-00055-of-00086.safetensors", + "model.layers.80.self_attn.k_proj.weight_scale": "model-00055-of-00086.safetensors", + "model.layers.80.self_attn.o_proj.weight": "model-00055-of-00086.safetensors", + "model.layers.80.self_attn.o_proj.weight_scale": "model-00055-of-00086.safetensors", + "model.layers.80.self_attn.q_proj.weight": "model-00055-of-00086.safetensors", + "model.layers.80.self_attn.q_proj.weight_scale": "model-00055-of-00086.safetensors", + "model.layers.80.self_attn.v_proj.weight": "model-00055-of-00086.safetensors", + "model.layers.80.self_attn.v_proj.weight_scale": "model-00055-of-00086.safetensors", + "model.layers.81.input_layernorm.weight": "model-00056-of-00086.safetensors", + "model.layers.81.mlp.down_proj.weight": "model-00056-of-00086.safetensors", + "model.layers.81.mlp.down_proj.weight_scale": "model-00056-of-00086.safetensors", + "model.layers.81.mlp.gate_proj.weight": "model-00056-of-00086.safetensors", + "model.layers.81.mlp.gate_proj.weight_scale": "model-00056-of-00086.safetensors", + "model.layers.81.mlp.up_proj.weight": "model-00056-of-00086.safetensors", + "model.layers.81.mlp.up_proj.weight_scale": "model-00056-of-00086.safetensors", + "model.layers.81.post_attention_layernorm.weight": "model-00056-of-00086.safetensors", + "model.layers.81.self_attn.k_proj.weight": "model-00055-of-00086.safetensors", + "model.layers.81.self_attn.k_proj.weight_scale": "model-00055-of-00086.safetensors", + "model.layers.81.self_attn.o_proj.weight": "model-00055-of-00086.safetensors", + "model.layers.81.self_attn.o_proj.weight_scale": "model-00055-of-00086.safetensors", + "model.layers.81.self_attn.q_proj.weight": "model-00055-of-00086.safetensors", + "model.layers.81.self_attn.q_proj.weight_scale": "model-00055-of-00086.safetensors", + "model.layers.81.self_attn.v_proj.weight": "model-00055-of-00086.safetensors", + "model.layers.81.self_attn.v_proj.weight_scale": "model-00055-of-00086.safetensors", + "model.layers.82.input_layernorm.weight": "model-00057-of-00086.safetensors", + "model.layers.82.mlp.down_proj.weight": "model-00057-of-00086.safetensors", + "model.layers.82.mlp.down_proj.weight_scale": "model-00057-of-00086.safetensors", + "model.layers.82.mlp.gate_proj.weight": "model-00056-of-00086.safetensors", + "model.layers.82.mlp.gate_proj.weight_scale": "model-00056-of-00086.safetensors", + "model.layers.82.mlp.up_proj.weight": "model-00056-of-00086.safetensors", + "model.layers.82.mlp.up_proj.weight_scale": "model-00056-of-00086.safetensors", + "model.layers.82.post_attention_layernorm.weight": "model-00057-of-00086.safetensors", + "model.layers.82.self_attn.k_proj.weight": "model-00056-of-00086.safetensors", + "model.layers.82.self_attn.k_proj.weight_scale": "model-00056-of-00086.safetensors", + "model.layers.82.self_attn.o_proj.weight": "model-00056-of-00086.safetensors", + "model.layers.82.self_attn.o_proj.weight_scale": "model-00056-of-00086.safetensors", + "model.layers.82.self_attn.q_proj.weight": "model-00056-of-00086.safetensors", + "model.layers.82.self_attn.q_proj.weight_scale": "model-00056-of-00086.safetensors", + "model.layers.82.self_attn.v_proj.weight": "model-00056-of-00086.safetensors", + "model.layers.82.self_attn.v_proj.weight_scale": "model-00056-of-00086.safetensors", + "model.layers.83.input_layernorm.weight": "model-00057-of-00086.safetensors", + "model.layers.83.mlp.down_proj.weight": "model-00057-of-00086.safetensors", + "model.layers.83.mlp.down_proj.weight_scale": "model-00057-of-00086.safetensors", + "model.layers.83.mlp.gate_proj.weight": "model-00057-of-00086.safetensors", + "model.layers.83.mlp.gate_proj.weight_scale": "model-00057-of-00086.safetensors", + "model.layers.83.mlp.up_proj.weight": "model-00057-of-00086.safetensors", + "model.layers.83.mlp.up_proj.weight_scale": "model-00057-of-00086.safetensors", + "model.layers.83.post_attention_layernorm.weight": "model-00057-of-00086.safetensors", + "model.layers.83.self_attn.k_proj.weight": "model-00057-of-00086.safetensors", + "model.layers.83.self_attn.k_proj.weight_scale": "model-00057-of-00086.safetensors", + "model.layers.83.self_attn.o_proj.weight": "model-00057-of-00086.safetensors", + "model.layers.83.self_attn.o_proj.weight_scale": "model-00057-of-00086.safetensors", + "model.layers.83.self_attn.q_proj.weight": "model-00057-of-00086.safetensors", + "model.layers.83.self_attn.q_proj.weight_scale": "model-00057-of-00086.safetensors", + "model.layers.83.self_attn.v_proj.weight": "model-00057-of-00086.safetensors", + "model.layers.83.self_attn.v_proj.weight_scale": "model-00057-of-00086.safetensors", + "model.layers.84.input_layernorm.weight": "model-00058-of-00086.safetensors", + "model.layers.84.mlp.down_proj.weight": "model-00058-of-00086.safetensors", + "model.layers.84.mlp.down_proj.weight_scale": "model-00058-of-00086.safetensors", + "model.layers.84.mlp.gate_proj.weight": "model-00058-of-00086.safetensors", + "model.layers.84.mlp.gate_proj.weight_scale": "model-00058-of-00086.safetensors", + "model.layers.84.mlp.up_proj.weight": "model-00058-of-00086.safetensors", + "model.layers.84.mlp.up_proj.weight_scale": "model-00058-of-00086.safetensors", + "model.layers.84.post_attention_layernorm.weight": "model-00058-of-00086.safetensors", + "model.layers.84.self_attn.k_proj.weight": "model-00057-of-00086.safetensors", + "model.layers.84.self_attn.k_proj.weight_scale": "model-00057-of-00086.safetensors", + "model.layers.84.self_attn.o_proj.weight": "model-00057-of-00086.safetensors", + "model.layers.84.self_attn.o_proj.weight_scale": "model-00057-of-00086.safetensors", + "model.layers.84.self_attn.q_proj.weight": "model-00057-of-00086.safetensors", + "model.layers.84.self_attn.q_proj.weight_scale": "model-00057-of-00086.safetensors", + "model.layers.84.self_attn.v_proj.weight": "model-00057-of-00086.safetensors", + "model.layers.84.self_attn.v_proj.weight_scale": "model-00057-of-00086.safetensors", + "model.layers.85.input_layernorm.weight": "model-00059-of-00086.safetensors", + "model.layers.85.mlp.down_proj.weight": "model-00059-of-00086.safetensors", + "model.layers.85.mlp.down_proj.weight_scale": "model-00059-of-00086.safetensors", + "model.layers.85.mlp.gate_proj.weight": "model-00058-of-00086.safetensors", + "model.layers.85.mlp.gate_proj.weight_scale": "model-00058-of-00086.safetensors", + "model.layers.85.mlp.up_proj.weight": "model-00058-of-00086.safetensors", + "model.layers.85.mlp.up_proj.weight_scale": "model-00058-of-00086.safetensors", + "model.layers.85.post_attention_layernorm.weight": "model-00059-of-00086.safetensors", + "model.layers.85.self_attn.k_proj.weight": "model-00058-of-00086.safetensors", + "model.layers.85.self_attn.k_proj.weight_scale": "model-00058-of-00086.safetensors", + "model.layers.85.self_attn.o_proj.weight": "model-00058-of-00086.safetensors", + "model.layers.85.self_attn.o_proj.weight_scale": "model-00058-of-00086.safetensors", + "model.layers.85.self_attn.q_proj.weight": "model-00058-of-00086.safetensors", + "model.layers.85.self_attn.q_proj.weight_scale": "model-00058-of-00086.safetensors", + "model.layers.85.self_attn.v_proj.weight": "model-00058-of-00086.safetensors", + "model.layers.85.self_attn.v_proj.weight_scale": "model-00058-of-00086.safetensors", + "model.layers.86.input_layernorm.weight": "model-00059-of-00086.safetensors", + "model.layers.86.mlp.down_proj.weight": "model-00059-of-00086.safetensors", + "model.layers.86.mlp.down_proj.weight_scale": "model-00059-of-00086.safetensors", + "model.layers.86.mlp.gate_proj.weight": "model-00059-of-00086.safetensors", + "model.layers.86.mlp.gate_proj.weight_scale": "model-00059-of-00086.safetensors", + "model.layers.86.mlp.up_proj.weight": "model-00059-of-00086.safetensors", + "model.layers.86.mlp.up_proj.weight_scale": "model-00059-of-00086.safetensors", + "model.layers.86.post_attention_layernorm.weight": "model-00059-of-00086.safetensors", + "model.layers.86.self_attn.k_proj.weight": "model-00059-of-00086.safetensors", + "model.layers.86.self_attn.k_proj.weight_scale": "model-00059-of-00086.safetensors", + "model.layers.86.self_attn.o_proj.weight": "model-00059-of-00086.safetensors", + "model.layers.86.self_attn.o_proj.weight_scale": "model-00059-of-00086.safetensors", + "model.layers.86.self_attn.q_proj.weight": "model-00059-of-00086.safetensors", + "model.layers.86.self_attn.q_proj.weight_scale": "model-00059-of-00086.safetensors", + "model.layers.86.self_attn.v_proj.weight": "model-00059-of-00086.safetensors", + "model.layers.86.self_attn.v_proj.weight_scale": "model-00059-of-00086.safetensors", + "model.layers.87.input_layernorm.weight": "model-00060-of-00086.safetensors", + "model.layers.87.mlp.down_proj.weight": "model-00060-of-00086.safetensors", + "model.layers.87.mlp.down_proj.weight_scale": "model-00060-of-00086.safetensors", + "model.layers.87.mlp.gate_proj.weight": "model-00060-of-00086.safetensors", + "model.layers.87.mlp.gate_proj.weight_scale": "model-00060-of-00086.safetensors", + "model.layers.87.mlp.up_proj.weight": "model-00060-of-00086.safetensors", + "model.layers.87.mlp.up_proj.weight_scale": "model-00060-of-00086.safetensors", + "model.layers.87.post_attention_layernorm.weight": "model-00060-of-00086.safetensors", + "model.layers.87.self_attn.k_proj.weight": "model-00059-of-00086.safetensors", + "model.layers.87.self_attn.k_proj.weight_scale": "model-00059-of-00086.safetensors", + "model.layers.87.self_attn.o_proj.weight": "model-00059-of-00086.safetensors", + "model.layers.87.self_attn.o_proj.weight_scale": "model-00059-of-00086.safetensors", + "model.layers.87.self_attn.q_proj.weight": "model-00059-of-00086.safetensors", + "model.layers.87.self_attn.q_proj.weight_scale": "model-00059-of-00086.safetensors", + "model.layers.87.self_attn.v_proj.weight": "model-00059-of-00086.safetensors", + "model.layers.87.self_attn.v_proj.weight_scale": "model-00059-of-00086.safetensors", + "model.layers.88.input_layernorm.weight": "model-00061-of-00086.safetensors", + "model.layers.88.mlp.down_proj.weight": "model-00061-of-00086.safetensors", + "model.layers.88.mlp.down_proj.weight_scale": "model-00061-of-00086.safetensors", + "model.layers.88.mlp.gate_proj.weight": "model-00060-of-00086.safetensors", + "model.layers.88.mlp.gate_proj.weight_scale": "model-00060-of-00086.safetensors", + "model.layers.88.mlp.up_proj.weight": "model-00060-of-00086.safetensors", + "model.layers.88.mlp.up_proj.weight_scale": "model-00060-of-00086.safetensors", + "model.layers.88.post_attention_layernorm.weight": "model-00061-of-00086.safetensors", + "model.layers.88.self_attn.k_proj.weight": "model-00060-of-00086.safetensors", + "model.layers.88.self_attn.k_proj.weight_scale": "model-00060-of-00086.safetensors", + "model.layers.88.self_attn.o_proj.weight": "model-00060-of-00086.safetensors", + "model.layers.88.self_attn.o_proj.weight_scale": "model-00060-of-00086.safetensors", + "model.layers.88.self_attn.q_proj.weight": "model-00060-of-00086.safetensors", + "model.layers.88.self_attn.q_proj.weight_scale": "model-00060-of-00086.safetensors", + "model.layers.88.self_attn.v_proj.weight": "model-00060-of-00086.safetensors", + "model.layers.88.self_attn.v_proj.weight_scale": "model-00060-of-00086.safetensors", + "model.layers.89.input_layernorm.weight": "model-00061-of-00086.safetensors", + "model.layers.89.mlp.down_proj.weight": "model-00061-of-00086.safetensors", + "model.layers.89.mlp.down_proj.weight_scale": "model-00061-of-00086.safetensors", + "model.layers.89.mlp.gate_proj.weight": "model-00061-of-00086.safetensors", + "model.layers.89.mlp.gate_proj.weight_scale": "model-00061-of-00086.safetensors", + "model.layers.89.mlp.up_proj.weight": "model-00061-of-00086.safetensors", + "model.layers.89.mlp.up_proj.weight_scale": "model-00061-of-00086.safetensors", + "model.layers.89.post_attention_layernorm.weight": "model-00061-of-00086.safetensors", + "model.layers.89.self_attn.k_proj.weight": "model-00061-of-00086.safetensors", + "model.layers.89.self_attn.k_proj.weight_scale": "model-00061-of-00086.safetensors", + "model.layers.89.self_attn.o_proj.weight": "model-00061-of-00086.safetensors", + "model.layers.89.self_attn.o_proj.weight_scale": "model-00061-of-00086.safetensors", + "model.layers.89.self_attn.q_proj.weight": "model-00061-of-00086.safetensors", + "model.layers.89.self_attn.q_proj.weight_scale": "model-00061-of-00086.safetensors", + "model.layers.89.self_attn.v_proj.weight": "model-00061-of-00086.safetensors", + "model.layers.89.self_attn.v_proj.weight_scale": "model-00061-of-00086.safetensors", + "model.layers.9.input_layernorm.weight": "model-00008-of-00086.safetensors", + "model.layers.9.mlp.down_proj.weight": "model-00008-of-00086.safetensors", + "model.layers.9.mlp.down_proj.weight_scale": "model-00008-of-00086.safetensors", + "model.layers.9.mlp.gate_proj.weight": "model-00008-of-00086.safetensors", + "model.layers.9.mlp.gate_proj.weight_scale": "model-00008-of-00086.safetensors", + "model.layers.9.mlp.up_proj.weight": "model-00008-of-00086.safetensors", + "model.layers.9.mlp.up_proj.weight_scale": "model-00008-of-00086.safetensors", + "model.layers.9.post_attention_layernorm.weight": "model-00008-of-00086.safetensors", + "model.layers.9.self_attn.k_proj.weight": "model-00007-of-00086.safetensors", + "model.layers.9.self_attn.k_proj.weight_scale": "model-00007-of-00086.safetensors", + "model.layers.9.self_attn.o_proj.weight": "model-00007-of-00086.safetensors", + "model.layers.9.self_attn.o_proj.weight_scale": "model-00007-of-00086.safetensors", + "model.layers.9.self_attn.q_proj.weight": "model-00007-of-00086.safetensors", + "model.layers.9.self_attn.q_proj.weight_scale": "model-00007-of-00086.safetensors", + "model.layers.9.self_attn.v_proj.weight": "model-00007-of-00086.safetensors", + "model.layers.9.self_attn.v_proj.weight_scale": "model-00007-of-00086.safetensors", + "model.layers.90.input_layernorm.weight": "model-00062-of-00086.safetensors", + "model.layers.90.mlp.down_proj.weight": "model-00062-of-00086.safetensors", + "model.layers.90.mlp.down_proj.weight_scale": "model-00062-of-00086.safetensors", + "model.layers.90.mlp.gate_proj.weight": "model-00062-of-00086.safetensors", + "model.layers.90.mlp.gate_proj.weight_scale": "model-00062-of-00086.safetensors", + "model.layers.90.mlp.up_proj.weight": "model-00062-of-00086.safetensors", + "model.layers.90.mlp.up_proj.weight_scale": "model-00062-of-00086.safetensors", + "model.layers.90.post_attention_layernorm.weight": "model-00062-of-00086.safetensors", + "model.layers.90.self_attn.k_proj.weight": "model-00061-of-00086.safetensors", + "model.layers.90.self_attn.k_proj.weight_scale": "model-00061-of-00086.safetensors", + "model.layers.90.self_attn.o_proj.weight": "model-00061-of-00086.safetensors", + "model.layers.90.self_attn.o_proj.weight_scale": "model-00061-of-00086.safetensors", + "model.layers.90.self_attn.q_proj.weight": "model-00061-of-00086.safetensors", + "model.layers.90.self_attn.q_proj.weight_scale": "model-00061-of-00086.safetensors", + "model.layers.90.self_attn.v_proj.weight": "model-00061-of-00086.safetensors", + "model.layers.90.self_attn.v_proj.weight_scale": "model-00061-of-00086.safetensors", + "model.layers.91.input_layernorm.weight": "model-00063-of-00086.safetensors", + "model.layers.91.mlp.down_proj.weight": "model-00063-of-00086.safetensors", + "model.layers.91.mlp.down_proj.weight_scale": "model-00063-of-00086.safetensors", + "model.layers.91.mlp.gate_proj.weight": "model-00062-of-00086.safetensors", + "model.layers.91.mlp.gate_proj.weight_scale": "model-00062-of-00086.safetensors", + "model.layers.91.mlp.up_proj.weight": "model-00062-of-00086.safetensors", + "model.layers.91.mlp.up_proj.weight_scale": "model-00062-of-00086.safetensors", + "model.layers.91.post_attention_layernorm.weight": "model-00063-of-00086.safetensors", + "model.layers.91.self_attn.k_proj.weight": "model-00062-of-00086.safetensors", + "model.layers.91.self_attn.k_proj.weight_scale": "model-00062-of-00086.safetensors", + "model.layers.91.self_attn.o_proj.weight": "model-00062-of-00086.safetensors", + "model.layers.91.self_attn.o_proj.weight_scale": "model-00062-of-00086.safetensors", + "model.layers.91.self_attn.q_proj.weight": "model-00062-of-00086.safetensors", + "model.layers.91.self_attn.q_proj.weight_scale": "model-00062-of-00086.safetensors", + "model.layers.91.self_attn.v_proj.weight": "model-00062-of-00086.safetensors", + "model.layers.91.self_attn.v_proj.weight_scale": "model-00062-of-00086.safetensors", + "model.layers.92.input_layernorm.weight": "model-00063-of-00086.safetensors", + "model.layers.92.mlp.down_proj.weight": "model-00063-of-00086.safetensors", + "model.layers.92.mlp.down_proj.weight_scale": "model-00063-of-00086.safetensors", + "model.layers.92.mlp.gate_proj.weight": "model-00063-of-00086.safetensors", + "model.layers.92.mlp.gate_proj.weight_scale": "model-00063-of-00086.safetensors", + "model.layers.92.mlp.up_proj.weight": "model-00063-of-00086.safetensors", + "model.layers.92.mlp.up_proj.weight_scale": "model-00063-of-00086.safetensors", + "model.layers.92.post_attention_layernorm.weight": "model-00063-of-00086.safetensors", + "model.layers.92.self_attn.k_proj.weight": "model-00063-of-00086.safetensors", + "model.layers.92.self_attn.k_proj.weight_scale": "model-00063-of-00086.safetensors", + "model.layers.92.self_attn.o_proj.weight": "model-00063-of-00086.safetensors", + "model.layers.92.self_attn.o_proj.weight_scale": "model-00063-of-00086.safetensors", + "model.layers.92.self_attn.q_proj.weight": "model-00063-of-00086.safetensors", + "model.layers.92.self_attn.q_proj.weight_scale": "model-00063-of-00086.safetensors", + "model.layers.92.self_attn.v_proj.weight": "model-00063-of-00086.safetensors", + "model.layers.92.self_attn.v_proj.weight_scale": "model-00063-of-00086.safetensors", + "model.layers.93.input_layernorm.weight": "model-00064-of-00086.safetensors", + "model.layers.93.mlp.down_proj.weight": "model-00064-of-00086.safetensors", + "model.layers.93.mlp.down_proj.weight_scale": "model-00064-of-00086.safetensors", + "model.layers.93.mlp.gate_proj.weight": "model-00064-of-00086.safetensors", + "model.layers.93.mlp.gate_proj.weight_scale": "model-00064-of-00086.safetensors", + "model.layers.93.mlp.up_proj.weight": "model-00064-of-00086.safetensors", + "model.layers.93.mlp.up_proj.weight_scale": "model-00064-of-00086.safetensors", + "model.layers.93.post_attention_layernorm.weight": "model-00064-of-00086.safetensors", + "model.layers.93.self_attn.k_proj.weight": "model-00063-of-00086.safetensors", + "model.layers.93.self_attn.k_proj.weight_scale": "model-00063-of-00086.safetensors", + "model.layers.93.self_attn.o_proj.weight": "model-00063-of-00086.safetensors", + "model.layers.93.self_attn.o_proj.weight_scale": "model-00063-of-00086.safetensors", + "model.layers.93.self_attn.q_proj.weight": "model-00063-of-00086.safetensors", + "model.layers.93.self_attn.q_proj.weight_scale": "model-00063-of-00086.safetensors", + "model.layers.93.self_attn.v_proj.weight": "model-00063-of-00086.safetensors", + "model.layers.93.self_attn.v_proj.weight_scale": "model-00063-of-00086.safetensors", + "model.layers.94.input_layernorm.weight": "model-00065-of-00086.safetensors", + "model.layers.94.mlp.down_proj.weight": "model-00065-of-00086.safetensors", + "model.layers.94.mlp.down_proj.weight_scale": "model-00065-of-00086.safetensors", + "model.layers.94.mlp.gate_proj.weight": "model-00064-of-00086.safetensors", + "model.layers.94.mlp.gate_proj.weight_scale": "model-00064-of-00086.safetensors", + "model.layers.94.mlp.up_proj.weight": "model-00064-of-00086.safetensors", + "model.layers.94.mlp.up_proj.weight_scale": "model-00064-of-00086.safetensors", + "model.layers.94.post_attention_layernorm.weight": "model-00065-of-00086.safetensors", + "model.layers.94.self_attn.k_proj.weight": "model-00064-of-00086.safetensors", + "model.layers.94.self_attn.k_proj.weight_scale": "model-00064-of-00086.safetensors", + "model.layers.94.self_attn.o_proj.weight": "model-00064-of-00086.safetensors", + "model.layers.94.self_attn.o_proj.weight_scale": "model-00064-of-00086.safetensors", + "model.layers.94.self_attn.q_proj.weight": "model-00064-of-00086.safetensors", + "model.layers.94.self_attn.q_proj.weight_scale": "model-00064-of-00086.safetensors", + "model.layers.94.self_attn.v_proj.weight": "model-00064-of-00086.safetensors", + "model.layers.94.self_attn.v_proj.weight_scale": "model-00064-of-00086.safetensors", + "model.layers.95.input_layernorm.weight": "model-00065-of-00086.safetensors", + "model.layers.95.mlp.down_proj.weight": "model-00065-of-00086.safetensors", + "model.layers.95.mlp.down_proj.weight_scale": "model-00065-of-00086.safetensors", + "model.layers.95.mlp.gate_proj.weight": "model-00065-of-00086.safetensors", + "model.layers.95.mlp.gate_proj.weight_scale": "model-00065-of-00086.safetensors", + "model.layers.95.mlp.up_proj.weight": "model-00065-of-00086.safetensors", + "model.layers.95.mlp.up_proj.weight_scale": "model-00065-of-00086.safetensors", + "model.layers.95.post_attention_layernorm.weight": "model-00065-of-00086.safetensors", + "model.layers.95.self_attn.k_proj.weight": "model-00065-of-00086.safetensors", + "model.layers.95.self_attn.k_proj.weight_scale": "model-00065-of-00086.safetensors", + "model.layers.95.self_attn.o_proj.weight": "model-00065-of-00086.safetensors", + "model.layers.95.self_attn.o_proj.weight_scale": "model-00065-of-00086.safetensors", + "model.layers.95.self_attn.q_proj.weight": "model-00065-of-00086.safetensors", + "model.layers.95.self_attn.q_proj.weight_scale": "model-00065-of-00086.safetensors", + "model.layers.95.self_attn.v_proj.weight": "model-00065-of-00086.safetensors", + "model.layers.95.self_attn.v_proj.weight_scale": "model-00065-of-00086.safetensors", + "model.layers.96.input_layernorm.weight": "model-00066-of-00086.safetensors", + "model.layers.96.mlp.down_proj.weight": "model-00066-of-00086.safetensors", + "model.layers.96.mlp.down_proj.weight_scale": "model-00066-of-00086.safetensors", + "model.layers.96.mlp.gate_proj.weight": "model-00066-of-00086.safetensors", + "model.layers.96.mlp.gate_proj.weight_scale": "model-00066-of-00086.safetensors", + "model.layers.96.mlp.up_proj.weight": "model-00066-of-00086.safetensors", + "model.layers.96.mlp.up_proj.weight_scale": "model-00066-of-00086.safetensors", + "model.layers.96.post_attention_layernorm.weight": "model-00066-of-00086.safetensors", + "model.layers.96.self_attn.k_proj.weight": "model-00065-of-00086.safetensors", + "model.layers.96.self_attn.k_proj.weight_scale": "model-00065-of-00086.safetensors", + "model.layers.96.self_attn.o_proj.weight": "model-00065-of-00086.safetensors", + "model.layers.96.self_attn.o_proj.weight_scale": "model-00065-of-00086.safetensors", + "model.layers.96.self_attn.q_proj.weight": "model-00065-of-00086.safetensors", + "model.layers.96.self_attn.q_proj.weight_scale": "model-00065-of-00086.safetensors", + "model.layers.96.self_attn.v_proj.weight": "model-00065-of-00086.safetensors", + "model.layers.96.self_attn.v_proj.weight_scale": "model-00065-of-00086.safetensors", + "model.layers.97.input_layernorm.weight": "model-00067-of-00086.safetensors", + "model.layers.97.mlp.down_proj.weight": "model-00067-of-00086.safetensors", + "model.layers.97.mlp.down_proj.weight_scale": "model-00067-of-00086.safetensors", + "model.layers.97.mlp.gate_proj.weight": "model-00066-of-00086.safetensors", + "model.layers.97.mlp.gate_proj.weight_scale": "model-00066-of-00086.safetensors", + "model.layers.97.mlp.up_proj.weight": "model-00066-of-00086.safetensors", + "model.layers.97.mlp.up_proj.weight_scale": "model-00066-of-00086.safetensors", + "model.layers.97.post_attention_layernorm.weight": "model-00067-of-00086.safetensors", + "model.layers.97.self_attn.k_proj.weight": "model-00066-of-00086.safetensors", + "model.layers.97.self_attn.k_proj.weight_scale": "model-00066-of-00086.safetensors", + "model.layers.97.self_attn.o_proj.weight": "model-00066-of-00086.safetensors", + "model.layers.97.self_attn.o_proj.weight_scale": "model-00066-of-00086.safetensors", + "model.layers.97.self_attn.q_proj.weight": "model-00066-of-00086.safetensors", + "model.layers.97.self_attn.q_proj.weight_scale": "model-00066-of-00086.safetensors", + "model.layers.97.self_attn.v_proj.weight": "model-00066-of-00086.safetensors", + "model.layers.97.self_attn.v_proj.weight_scale": "model-00066-of-00086.safetensors", + "model.layers.98.input_layernorm.weight": "model-00067-of-00086.safetensors", + "model.layers.98.mlp.down_proj.weight": "model-00067-of-00086.safetensors", + "model.layers.98.mlp.down_proj.weight_scale": "model-00067-of-00086.safetensors", + "model.layers.98.mlp.gate_proj.weight": "model-00067-of-00086.safetensors", + "model.layers.98.mlp.gate_proj.weight_scale": "model-00067-of-00086.safetensors", + "model.layers.98.mlp.up_proj.weight": "model-00067-of-00086.safetensors", + "model.layers.98.mlp.up_proj.weight_scale": "model-00067-of-00086.safetensors", + "model.layers.98.post_attention_layernorm.weight": "model-00067-of-00086.safetensors", + "model.layers.98.self_attn.k_proj.weight": "model-00067-of-00086.safetensors", + "model.layers.98.self_attn.k_proj.weight_scale": "model-00067-of-00086.safetensors", + "model.layers.98.self_attn.o_proj.weight": "model-00067-of-00086.safetensors", + "model.layers.98.self_attn.o_proj.weight_scale": "model-00067-of-00086.safetensors", + "model.layers.98.self_attn.q_proj.weight": "model-00067-of-00086.safetensors", + "model.layers.98.self_attn.q_proj.weight_scale": "model-00067-of-00086.safetensors", + "model.layers.98.self_attn.v_proj.weight": "model-00067-of-00086.safetensors", + "model.layers.98.self_attn.v_proj.weight_scale": "model-00067-of-00086.safetensors", + "model.layers.99.input_layernorm.weight": "model-00068-of-00086.safetensors", + "model.layers.99.mlp.down_proj.weight": "model-00068-of-00086.safetensors", + "model.layers.99.mlp.down_proj.weight_scale": "model-00068-of-00086.safetensors", + "model.layers.99.mlp.gate_proj.weight": "model-00068-of-00086.safetensors", + "model.layers.99.mlp.gate_proj.weight_scale": "model-00068-of-00086.safetensors", + "model.layers.99.mlp.up_proj.weight": "model-00068-of-00086.safetensors", + "model.layers.99.mlp.up_proj.weight_scale": "model-00068-of-00086.safetensors", + "model.layers.99.post_attention_layernorm.weight": "model-00068-of-00086.safetensors", + "model.layers.99.self_attn.k_proj.weight": "model-00067-of-00086.safetensors", + "model.layers.99.self_attn.k_proj.weight_scale": "model-00067-of-00086.safetensors", + "model.layers.99.self_attn.o_proj.weight": "model-00067-of-00086.safetensors", + "model.layers.99.self_attn.o_proj.weight_scale": "model-00067-of-00086.safetensors", + "model.layers.99.self_attn.q_proj.weight": "model-00067-of-00086.safetensors", + "model.layers.99.self_attn.q_proj.weight_scale": "model-00067-of-00086.safetensors", + "model.layers.99.self_attn.v_proj.weight": "model-00067-of-00086.safetensors", + "model.layers.99.self_attn.v_proj.weight_scale": "model-00067-of-00086.safetensors", + "model.norm.weight": "model-00085-of-00086.safetensors" + } +} diff --git a/recipe.yaml b/recipe.yaml new file mode 100644 index 0000000000000000000000000000000000000000..de3fa93d9022d0ae737469d25310fecdf6d9126b --- /dev/null +++ b/recipe.yaml @@ -0,0 +1,7 @@ +default_stage: + default_modifiers: + SmoothQuantModifier: {smoothing_strength: 0.8} + GPTQModifier: + scheme: W8A8 + targets: Linear + ignore: [lm_head] diff --git a/special_tokens_map.json b/special_tokens_map.json new file mode 100644 index 0000000000000000000000000000000000000000..344c8261025248cbe380e52f8730a03149d599e1 --- /dev/null +++ b/special_tokens_map.json @@ -0,0 +1,23 @@ +{ + "bos_token": { + "content": "<|begin_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + }, + "eos_token": { + "content": "<|eot_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + }, + "pad_token": { + "content": "<|eot_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + } +} diff --git a/tokenizer.json b/tokenizer.json new file mode 100644 index 0000000000000000000000000000000000000000..bf19ce7e907d493e143806468b7d256e6a33e6dd --- /dev/null +++ b/tokenizer.json @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:52716f60c3ad328509fa37cdded9a2f1196ecae463f5480f5d38c66a25e7a7dc +size 17210019 diff --git a/tokenizer_config.json b/tokenizer_config.json new file mode 100644 index 0000000000000000000000000000000000000000..2c7095124f783bd0c1ab751763abd2b544bc5cda --- /dev/null +++ b/tokenizer_config.json @@ -0,0 +1,2067 @@ +{ + "added_tokens_decoder": { + "128000": { + "content": "<|begin_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128001": { + "content": "<|end_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128002": { + "content": "<|reserved_special_token_0|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128003": { + "content": "<|reserved_special_token_1|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128004": { + "content": "<|finetune_right_pad_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128005": { + "content": "<|reserved_special_token_2|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128006": { + "content": "<|start_header_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128007": { + "content": "<|end_header_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128008": { + "content": "<|eom_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128009": { + "content": "<|eot_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128010": { + "content": "<|python_tag|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128011": { + "content": "<|reserved_special_token_3|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128012": { + "content": "<|reserved_special_token_4|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128013": { + "content": "<|reserved_special_token_5|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128014": { + "content": "<|reserved_special_token_6|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128015": { + "content": "<|reserved_special_token_7|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128016": { + "content": "<|reserved_special_token_8|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128017": { + "content": "<|reserved_special_token_9|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128018": { + "content": "<|reserved_special_token_10|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128019": { + "content": "<|reserved_special_token_11|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128020": { + "content": "<|reserved_special_token_12|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128021": { + "content": "<|reserved_special_token_13|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128022": { + "content": "<|reserved_special_token_14|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128023": { + "content": "<|reserved_special_token_15|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128024": { + "content": "<|reserved_special_token_16|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128025": { + "content": "<|reserved_special_token_17|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128026": { + "content": "<|reserved_special_token_18|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128027": { + "content": "<|reserved_special_token_19|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128028": { + "content": "<|reserved_special_token_20|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128029": { + "content": "<|reserved_special_token_21|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128030": { + "content": "<|reserved_special_token_22|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128031": { + "content": "<|reserved_special_token_23|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128032": { + "content": "<|reserved_special_token_24|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128033": { + "content": "<|reserved_special_token_25|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128034": { + "content": "<|reserved_special_token_26|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128035": { + "content": "<|reserved_special_token_27|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128036": { + "content": "<|reserved_special_token_28|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128037": { + "content": "<|reserved_special_token_29|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128038": { + "content": "<|reserved_special_token_30|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128039": { + "content": "<|reserved_special_token_31|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128040": { + "content": "<|reserved_special_token_32|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128041": { + "content": "<|reserved_special_token_33|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128042": { + "content": "<|reserved_special_token_34|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128043": { + "content": "<|reserved_special_token_35|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128044": { + "content": "<|reserved_special_token_36|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128045": { + "content": "<|reserved_special_token_37|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128046": { + "content": "<|reserved_special_token_38|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128047": { + "content": "<|reserved_special_token_39|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128048": { + "content": "<|reserved_special_token_40|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128049": { + "content": "<|reserved_special_token_41|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128050": { + "content": "<|reserved_special_token_42|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128051": { + "content": "<|reserved_special_token_43|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128052": { + "content": "<|reserved_special_token_44|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128053": { + "content": "<|reserved_special_token_45|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128054": { + "content": "<|reserved_special_token_46|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128055": { + "content": "<|reserved_special_token_47|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128056": { + "content": "<|reserved_special_token_48|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128057": { + "content": "<|reserved_special_token_49|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128058": { + "content": "<|reserved_special_token_50|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128059": { + "content": "<|reserved_special_token_51|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128060": { + "content": "<|reserved_special_token_52|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128061": { + "content": "<|reserved_special_token_53|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128062": { + "content": "<|reserved_special_token_54|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128063": { + "content": "<|reserved_special_token_55|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128064": { + "content": "<|reserved_special_token_56|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128065": { + "content": "<|reserved_special_token_57|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128066": { + "content": "<|reserved_special_token_58|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128067": { + "content": "<|reserved_special_token_59|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128068": { + "content": "<|reserved_special_token_60|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128069": { + "content": "<|reserved_special_token_61|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128070": { + "content": "<|reserved_special_token_62|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128071": { + "content": "<|reserved_special_token_63|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128072": { + "content": "<|reserved_special_token_64|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128073": { + "content": "<|reserved_special_token_65|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128074": { + "content": "<|reserved_special_token_66|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128075": { + "content": "<|reserved_special_token_67|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128076": { + "content": "<|reserved_special_token_68|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128077": { + "content": "<|reserved_special_token_69|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128078": { + "content": "<|reserved_special_token_70|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128079": { + "content": "<|reserved_special_token_71|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128080": { + "content": "<|reserved_special_token_72|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128081": { + "content": "<|reserved_special_token_73|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128082": { + "content": "<|reserved_special_token_74|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128083": { + "content": "<|reserved_special_token_75|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128084": { + "content": "<|reserved_special_token_76|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128085": { + "content": "<|reserved_special_token_77|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128086": { + "content": "<|reserved_special_token_78|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128087": { + "content": "<|reserved_special_token_79|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128088": { + "content": "<|reserved_special_token_80|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128089": { + "content": "<|reserved_special_token_81|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128090": { + "content": "<|reserved_special_token_82|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128091": { + "content": "<|reserved_special_token_83|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128092": { + "content": "<|reserved_special_token_84|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128093": { + "content": "<|reserved_special_token_85|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128094": { + "content": "<|reserved_special_token_86|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128095": { + "content": "<|reserved_special_token_87|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128096": { + "content": "<|reserved_special_token_88|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128097": { + "content": "<|reserved_special_token_89|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128098": { + "content": "<|reserved_special_token_90|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128099": { + "content": "<|reserved_special_token_91|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128100": { + "content": "<|reserved_special_token_92|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128101": { + "content": "<|reserved_special_token_93|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128102": { + "content": "<|reserved_special_token_94|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128103": { + "content": "<|reserved_special_token_95|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128104": { + "content": "<|reserved_special_token_96|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128105": { + "content": "<|reserved_special_token_97|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128106": { + "content": "<|reserved_special_token_98|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128107": { + "content": "<|reserved_special_token_99|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128108": { + "content": "<|reserved_special_token_100|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128109": { + "content": "<|reserved_special_token_101|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128110": { + "content": "<|reserved_special_token_102|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128111": { + "content": "<|reserved_special_token_103|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128112": { + "content": "<|reserved_special_token_104|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128113": { + "content": "<|reserved_special_token_105|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128114": { + "content": "<|reserved_special_token_106|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128115": { + "content": "<|reserved_special_token_107|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128116": { + "content": "<|reserved_special_token_108|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128117": { + "content": "<|reserved_special_token_109|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128118": { + "content": "<|reserved_special_token_110|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128119": { + "content": "<|reserved_special_token_111|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128120": { + "content": "<|reserved_special_token_112|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128121": { + "content": "<|reserved_special_token_113|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128122": { + "content": "<|reserved_special_token_114|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128123": { + "content": "<|reserved_special_token_115|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128124": { + "content": "<|reserved_special_token_116|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128125": { + "content": "<|reserved_special_token_117|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128126": { + "content": "<|reserved_special_token_118|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128127": { + "content": "<|reserved_special_token_119|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128128": { + "content": "<|reserved_special_token_120|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128129": { + "content": "<|reserved_special_token_121|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128130": { + "content": "<|reserved_special_token_122|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128131": { + "content": "<|reserved_special_token_123|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128132": { + "content": "<|reserved_special_token_124|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128133": { + "content": "<|reserved_special_token_125|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128134": { + "content": "<|reserved_special_token_126|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128135": { + "content": "<|reserved_special_token_127|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128136": { + "content": "<|reserved_special_token_128|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128137": { + "content": "<|reserved_special_token_129|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128138": { + "content": "<|reserved_special_token_130|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128139": { + "content": "<|reserved_special_token_131|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128140": { + "content": "<|reserved_special_token_132|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128141": { + "content": "<|reserved_special_token_133|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128142": { + "content": "<|reserved_special_token_134|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128143": { + "content": "<|reserved_special_token_135|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128144": { + "content": "<|reserved_special_token_136|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128145": { + "content": "<|reserved_special_token_137|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128146": { + "content": "<|reserved_special_token_138|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128147": { + "content": "<|reserved_special_token_139|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128148": { + "content": "<|reserved_special_token_140|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128149": { + "content": "<|reserved_special_token_141|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128150": { + "content": "<|reserved_special_token_142|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128151": { + "content": "<|reserved_special_token_143|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128152": { + "content": "<|reserved_special_token_144|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128153": { + "content": "<|reserved_special_token_145|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128154": { + "content": "<|reserved_special_token_146|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128155": { + "content": "<|reserved_special_token_147|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128156": { + "content": "<|reserved_special_token_148|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128157": { + "content": "<|reserved_special_token_149|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128158": { + "content": "<|reserved_special_token_150|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128159": { + "content": "<|reserved_special_token_151|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128160": { + "content": "<|reserved_special_token_152|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128161": { + "content": "<|reserved_special_token_153|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128162": { + "content": "<|reserved_special_token_154|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128163": { + "content": "<|reserved_special_token_155|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128164": { + "content": "<|reserved_special_token_156|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128165": { + "content": "<|reserved_special_token_157|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128166": { + "content": "<|reserved_special_token_158|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128167": { + "content": "<|reserved_special_token_159|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128168": { + "content": "<|reserved_special_token_160|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128169": { + "content": "<|reserved_special_token_161|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128170": { + "content": "<|reserved_special_token_162|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128171": { + "content": "<|reserved_special_token_163|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128172": { + "content": "<|reserved_special_token_164|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128173": { + "content": "<|reserved_special_token_165|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128174": { + "content": "<|reserved_special_token_166|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128175": { + "content": "<|reserved_special_token_167|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128176": { + "content": "<|reserved_special_token_168|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128177": { + "content": "<|reserved_special_token_169|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128178": { + "content": "<|reserved_special_token_170|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128179": { + "content": "<|reserved_special_token_171|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128180": { + "content": "<|reserved_special_token_172|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128181": { + "content": "<|reserved_special_token_173|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128182": { + "content": "<|reserved_special_token_174|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128183": { + "content": "<|reserved_special_token_175|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128184": { + "content": "<|reserved_special_token_176|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128185": { + "content": "<|reserved_special_token_177|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128186": { + "content": "<|reserved_special_token_178|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128187": { + "content": "<|reserved_special_token_179|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128188": { + "content": "<|reserved_special_token_180|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128189": { + "content": "<|reserved_special_token_181|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128190": { + "content": "<|reserved_special_token_182|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128191": { + "content": "<|reserved_special_token_183|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128192": { + "content": "<|reserved_special_token_184|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128193": { + "content": "<|reserved_special_token_185|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128194": { + "content": "<|reserved_special_token_186|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128195": { + "content": "<|reserved_special_token_187|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128196": { + "content": "<|reserved_special_token_188|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128197": { + "content": "<|reserved_special_token_189|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128198": { + "content": "<|reserved_special_token_190|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128199": { + "content": "<|reserved_special_token_191|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128200": { + "content": "<|reserved_special_token_192|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128201": { + "content": "<|reserved_special_token_193|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128202": { + "content": "<|reserved_special_token_194|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128203": { + "content": "<|reserved_special_token_195|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128204": { + "content": "<|reserved_special_token_196|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128205": { + "content": "<|reserved_special_token_197|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128206": { + "content": "<|reserved_special_token_198|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128207": { + "content": "<|reserved_special_token_199|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128208": { + "content": "<|reserved_special_token_200|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128209": { + "content": "<|reserved_special_token_201|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128210": { + "content": "<|reserved_special_token_202|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128211": { + "content": "<|reserved_special_token_203|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128212": { + "content": "<|reserved_special_token_204|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128213": { + "content": "<|reserved_special_token_205|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128214": { + "content": "<|reserved_special_token_206|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128215": { + "content": "<|reserved_special_token_207|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128216": { + "content": "<|reserved_special_token_208|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128217": { + "content": "<|reserved_special_token_209|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128218": { + "content": "<|reserved_special_token_210|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128219": { + "content": "<|reserved_special_token_211|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128220": { + "content": "<|reserved_special_token_212|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128221": { + "content": "<|reserved_special_token_213|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128222": { + "content": "<|reserved_special_token_214|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128223": { + "content": "<|reserved_special_token_215|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128224": { + "content": "<|reserved_special_token_216|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128225": { + "content": "<|reserved_special_token_217|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128226": { + "content": "<|reserved_special_token_218|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128227": { + "content": "<|reserved_special_token_219|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128228": { + "content": "<|reserved_special_token_220|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128229": { + "content": "<|reserved_special_token_221|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128230": { + "content": "<|reserved_special_token_222|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128231": { + "content": "<|reserved_special_token_223|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128232": { + "content": "<|reserved_special_token_224|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128233": { + "content": "<|reserved_special_token_225|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128234": { + "content": "<|reserved_special_token_226|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128235": { + "content": "<|reserved_special_token_227|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128236": { + "content": "<|reserved_special_token_228|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128237": { + "content": "<|reserved_special_token_229|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128238": { + "content": "<|reserved_special_token_230|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128239": { + "content": "<|reserved_special_token_231|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128240": { + "content": "<|reserved_special_token_232|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128241": { + "content": "<|reserved_special_token_233|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128242": { + "content": "<|reserved_special_token_234|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128243": { + "content": "<|reserved_special_token_235|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128244": { + "content": "<|reserved_special_token_236|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128245": { + "content": "<|reserved_special_token_237|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128246": { + "content": "<|reserved_special_token_238|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128247": { + "content": "<|reserved_special_token_239|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128248": { + "content": "<|reserved_special_token_240|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128249": { + "content": "<|reserved_special_token_241|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128250": { + "content": "<|reserved_special_token_242|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128251": { + "content": "<|reserved_special_token_243|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128252": { + "content": "<|reserved_special_token_244|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128253": { + "content": "<|reserved_special_token_245|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128254": { + "content": "<|reserved_special_token_246|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128255": { + "content": "<|reserved_special_token_247|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + } + }, + "bos_token": "<|begin_of_text|>", + "clean_up_tokenization_spaces": true, + "eos_token": "<|eot_id|>", + "extra_special_tokens": {}, + "max_length": 4096, + "model_input_names": [ + "input_ids", + "attention_mask" + ], + "model_max_length": 131072, + "pad_token": "<|eot_id|>", + "stride": 0, + "tokenizer_class": "PreTrainedTokenizer", + "truncation_side": "right", + "truncation_strategy": "longest_first" +}