Training in progress, step 1650, checkpoint

.gitattributes

@@ -36,3 +36,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 tokenizer.json filter=lfs diff=lfs merge=lfs -text
 checkpoint-550/tokenizer.json filter=lfs diff=lfs merge=lfs -text
 checkpoint-1100/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+checkpoint-1650/tokenizer.json filter=lfs diff=lfs merge=lfs -text

checkpoint-1650/added_tokens.json

@@ -0,0 +1,24 @@
+{
+  "</tool_call>": 151658,
+  "<tool_call>": 151657,
+  "<|box_end|>": 151649,
+  "<|box_start|>": 151648,
+  "<|endoftext|>": 151643,
+  "<|file_sep|>": 151664,
+  "<|fim_middle|>": 151660,
+  "<|fim_pad|>": 151662,
+  "<|fim_prefix|>": 151659,
+  "<|fim_suffix|>": 151661,
+  "<|im_end|>": 151645,
+  "<|im_start|>": 151644,
+  "<|image_pad|>": 151655,
+  "<|object_ref_end|>": 151647,
+  "<|object_ref_start|>": 151646,
+  "<|quad_end|>": 151651,
+  "<|quad_start|>": 151650,
+  "<|repo_name|>": 151663,
+  "<|video_pad|>": 151656,
+  "<|vision_end|>": 151653,
+  "<|vision_pad|>": 151654,
+  "<|vision_start|>": 151652
+}

checkpoint-1650/config.json

@@ -0,0 +1,32 @@
+{
+  "_name_or_path": "Qwen7B-Math-L28-LP32",
+  "architectures": [
+    "Qwen2ForCausalLM"
+  ],
+  "attention_dropout": 0.0,
+  "bos_token_id": 151643,
+  "end_conv_idx": 16,
+  "eos_token_id": 151645,
+  "hidden_act": "silu",
+  "hidden_size": 3584,
+  "initializer_range": 0.02,
+  "intermediate_size": 18944,
+  "max_position_embeddings": 32768,
+  "max_window_layers": 28,
+  "model_type": "qwen2",
+  "num_attention_heads": 28,
+  "num_conv": 1,
+  "num_hidden_layers": 32,
+  "num_key_value_heads": 4,
+  "rms_norm_eps": 1e-06,
+  "rope_scaling": null,
+  "rope_theta": 300000.0,
+  "sliding_window": null,
+  "start_conv_idx": 12,
+  "tie_word_embeddings": false,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.49.0",
+  "use_cache": false,
+  "use_sliding_window": false,
+  "vocab_size": 152064
+}

checkpoint-1650/generation_config.json

@@ -0,0 +1,9 @@
+{
+  "bos_token_id": 151643,
+  "eos_token_id": [
+    151645,
+    151643
+  ],
+  "pad_token_id": 151643,
+  "transformers_version": "4.49.0"
+}

checkpoint-1650/global_step1650/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:27755436c6a3324ad6cad5329630b2873d1e666b04ef3f4c9c0c629b4046b135
+size 12821777702

checkpoint-1650/global_step1650/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2fe07ee3312fb86f96120c6a32b709eff8c3311f546ce6e621f448a227be000c
+size 12821777702

checkpoint-1650/global_step1650/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3d22ec558327df606cdac879a62ffd032d278e46c2b183a4043e9ca55df0381f
+size 12821777702

checkpoint-1650/global_step1650/bf16_zero_pp_rank_3_mp_rank_00_optim_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b8fcd752d59f4dd0002fb60cfc1755e059ffae19b30aaaabb620591a3f90c764
+size 12821777702

checkpoint-1650/global_step1650/bf16_zero_pp_rank_4_mp_rank_00_optim_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e81d3565da7e5b538fccf3bbbcb58c0cde44ed5c07f2bf4697a8c937ea443da0
+size 12821777702

checkpoint-1650/global_step1650/bf16_zero_pp_rank_5_mp_rank_00_optim_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f2dc5ce04b6c412ef59050b788a232f8cd3efc4c77220767fc7bdc20e79e1844
+size 12821777702

checkpoint-1650/global_step1650/bf16_zero_pp_rank_6_mp_rank_00_optim_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:457eada28f3fda1a4afbfd25d9510c37af6fc3dcf386a22718c8d7db4a5ab6ac
+size 12821777702

checkpoint-1650/global_step1650/bf16_zero_pp_rank_7_mp_rank_00_optim_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1491066678bbaa2e817b7e2c9a8661076e4db04b98ce45defc89142dd71a398a
+size 12821777702

checkpoint-1650/global_step1650/zero_pp_rank_0_mp_rank_00_model_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e481415f6d27f55d6624c19f8bb47fc4ea99432c01ef532ef701cd83678ad5a4
+size 188165

checkpoint-1650/global_step1650/zero_pp_rank_1_mp_rank_00_model_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cd9dcc8b2900d97f75b5ffb9e3fabb4ddca9f9b4f5285d6f7224535b2e418b55
+size 188165

checkpoint-1650/global_step1650/zero_pp_rank_2_mp_rank_00_model_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5ca227a1d6b5babec81bdbcfc4a8748a0508d31cc20a688e63908bd73ff2beb0
+size 188165

checkpoint-1650/global_step1650/zero_pp_rank_3_mp_rank_00_model_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:129174492e79c3d2a8bdb3cc9192244ff38075186c9ebcb1ab7cc392a3c050f1
+size 188165

checkpoint-1650/global_step1650/zero_pp_rank_4_mp_rank_00_model_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6b4e0b754d0f83579105fa9d80837718d724875d613f4f17bd706fd326cf8281
+size 188165

checkpoint-1650/global_step1650/zero_pp_rank_5_mp_rank_00_model_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f696d396a3f0381df10df15701ae424dbc1726b2048cf9b54b7a7aa3a54a6964
+size 188165

checkpoint-1650/global_step1650/zero_pp_rank_6_mp_rank_00_model_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fd2ff97898507370c076375a7d798cc5d4ab7b89d5d5be66ae9e6f073034b0fc
+size 188165

checkpoint-1650/global_step1650/zero_pp_rank_7_mp_rank_00_model_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2a77569dfb3c360ef61eacdef665361ad89034c889196f869dcdc566f284705b
+size 188165

checkpoint-1650/latest

@@ -0,0 +1 @@
+global_step1650

checkpoint-1650/model-00001-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cebe966a4dfbea40d3b13a7a60508a2a58fa0d15c0cf9fd94aee9e8d4666d9f1
+size 4877660776

checkpoint-1650/model-00002-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0569dfc6396a999fa7e0937b09ee530a3b7118e516690ef8b57825895cd071ea
+size 4932751008

checkpoint-1650/model-00003-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:09e028cc7d200ddd07ed56f3e6d8cb282ff1f7c3a70bb67649ddafb5803b00c3
+size 4991495888

checkpoint-1650/model-00004-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:38bd6134ca20ad8e1c5be08a5d1a1a755ff275b18e50bd542cfc8fa5b5915bf0
+size 2293832048

checkpoint-1650/model.safetensors.index.json

@@ -0,0 +1,394 @@
+{
+  "metadata": {
+    "total_size": 17095695360
+  },
+  "weight_map": {
+    "lm_head.weight": "model-00004-of-00004.safetensors",
+    "model.embed_tokens.weight": "model-00001-of-00004.safetensors",
+    "model.layers.0.input_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.0.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.0.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.0.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.0.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.0.self_attn.k_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.0.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.0.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.0.self_attn.q_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.0.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.0.self_attn.v_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.0.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.1.input_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.1.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.1.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.1.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.1.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.1.self_attn.k_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.1.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.1.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.1.self_attn.q_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.1.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.1.self_attn.v_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.1.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.10.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.10.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.10.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.10.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.10.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.10.self_attn.k_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.10.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.10.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.10.self_attn.q_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.10.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.10.self_attn.v_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.10.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.11.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.11.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.11.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.11.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.11.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.11.self_attn.k_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.11.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.11.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.11.self_attn.q_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.11.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.11.self_attn.v_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.11.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.12.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.12.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.12.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.12.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.12.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.12.self_attn.k_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.12.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.12.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.12.self_attn.q_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.12.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.12.self_attn.v_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.12.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.13.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.13.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.13.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.13.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.13.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.13.self_attn.k_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.13.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.13.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.13.self_attn.q_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.13.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.13.self_attn.v_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.13.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.14.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.14.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.14.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.14.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.14.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.14.self_attn.k_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.14.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.14.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.14.self_attn.q_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.14.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.14.self_attn.v_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.14.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.15.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.15.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.15.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.15.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.15.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.15.self_attn.k_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.15.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.15.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.15.self_attn.q_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.15.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.15.self_attn.v_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.15.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.16.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.16.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.16.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.16.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.16.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.16.self_attn.k_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.16.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.16.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.16.self_attn.q_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.16.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.16.self_attn.v_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.16.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.17.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.17.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.17.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.17.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.17.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.17.self_attn.k_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.17.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.17.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.17.self_attn.q_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.17.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.17.self_attn.v_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.17.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.18.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.18.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.18.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.18.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.18.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.18.self_attn.k_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.18.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.18.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.18.self_attn.q_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.18.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.18.self_attn.v_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.18.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.19.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.19.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.19.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.19.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.19.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.19.self_attn.k_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.19.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.19.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.19.self_attn.q_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.19.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.19.self_attn.v_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.19.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.2.input_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.2.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.2.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.2.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.2.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.2.self_attn.k_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.2.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.2.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.2.self_attn.q_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.2.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.2.self_attn.v_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.2.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.20.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.20.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.20.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.20.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.20.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.20.self_attn.k_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.20.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.20.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.20.self_attn.q_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.20.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.20.self_attn.v_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.20.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.21.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.21.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.21.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.21.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.21.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.21.self_attn.k_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.21.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.21.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.21.self_attn.q_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.21.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.21.self_attn.v_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.21.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.22.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.22.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.22.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.22.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.22.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.22.self_attn.k_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.22.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.22.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.22.self_attn.q_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.22.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.22.self_attn.v_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.22.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.23.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.23.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.23.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.23.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.23.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.23.self_attn.k_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.23.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.23.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.23.self_attn.q_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.23.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.23.self_attn.v_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.23.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.24.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.24.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.24.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.24.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.24.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.24.self_attn.k_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.24.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.24.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.24.self_attn.q_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.24.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.24.self_attn.v_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.24.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.25.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.25.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.25.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.25.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.25.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.25.self_attn.k_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.25.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.25.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.25.self_attn.q_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.25.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.25.self_attn.v_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.25.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.26.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.26.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.26.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.26.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.26.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.26.self_attn.k_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.26.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.26.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.26.self_attn.q_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.26.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.26.self_attn.v_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.26.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.27.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.27.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.27.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.27.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.27.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.27.self_attn.k_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.27.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.27.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.27.self_attn.q_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.27.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.27.self_attn.v_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.27.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.28.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.28.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.28.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.28.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.28.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.28.self_attn.k_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.28.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.28.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.28.self_attn.q_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.28.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.28.self_attn.v_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.28.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.29.input_layernorm.weight": "model-00004-of-00004.safetensors",
+    "model.layers.29.mlp.down_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.29.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.29.mlp.up_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.29.post_attention_layernorm.weight": "model-00004-of-00004.safetensors",
+    "model.layers.29.self_attn.k_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.29.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.29.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.29.self_attn.q_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.29.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.29.self_attn.v_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.29.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.3.input_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.3.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.3.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.3.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.3.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.3.self_attn.k_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.3.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.3.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.3.self_attn.q_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.3.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.3.self_attn.v_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.3.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.30.input_layernorm.weight": "model-00004-of-00004.safetensors",
+    "model.layers.30.mlp.down_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.30.mlp.gate_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.30.mlp.up_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.30.post_attention_layernorm.weight": "model-00004-of-00004.safetensors",
+    "model.layers.30.self_attn.k_proj.bias": "model-00004-of-00004.safetensors",
+    "model.layers.30.self_attn.k_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.30.self_attn.o_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.30.self_attn.q_proj.bias": "model-00004-of-00004.safetensors",
+    "model.layers.30.self_attn.q_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.30.self_attn.v_proj.bias": "model-00004-of-00004.safetensors",
+    "model.layers.30.self_attn.v_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.31.input_layernorm.weight": "model-00004-of-00004.safetensors",
+    "model.layers.31.mlp.down_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.31.mlp.gate_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.31.mlp.up_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.31.post_attention_layernorm.weight": "model-00004-of-00004.safetensors",
+    "model.layers.31.self_attn.k_proj.bias": "model-00004-of-00004.safetensors",
+    "model.layers.31.self_attn.k_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.31.self_attn.o_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.31.self_attn.q_proj.bias": "model-00004-of-00004.safetensors",
+    "model.layers.31.self_attn.q_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.31.self_attn.v_proj.bias": "model-00004-of-00004.safetensors",
+    "model.layers.31.self_attn.v_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.4.input_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.4.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.4.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.4.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.4.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.4.self_attn.k_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.4.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.4.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.4.self_attn.q_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.4.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.4.self_attn.v_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.4.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.5.input_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.5.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.5.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.5.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.5.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.5.self_attn.k_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.5.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.5.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.5.self_attn.q_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.5.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.5.self_attn.v_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.5.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.6.input_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.6.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.6.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.6.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.6.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.6.self_attn.k_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.6.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.6.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.6.self_attn.q_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.6.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.6.self_attn.v_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.6.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.7.input_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.7.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.7.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.7.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.7.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.7.self_attn.k_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.7.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.7.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.7.self_attn.q_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.7.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.7.self_attn.v_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.7.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.8.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.8.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.8.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.8.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.8.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.8.self_attn.k_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.8.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.8.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.8.self_attn.q_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.8.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.8.self_attn.v_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.8.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.9.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.9.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.9.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.9.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.9.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.9.self_attn.k_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.9.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.9.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.9.self_attn.q_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.9.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.9.self_attn.v_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.9.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "model.norm.weight": "model-00004-of-00004.safetensors"
+  }
+}

checkpoint-1650/rng_state_0.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ad8a35afd8967cbb748405387e44426e43ad127028e826eddc9b67d2ca873c85
+size 15984

checkpoint-1650/rng_state_1.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f338ce80d7c441076bfc8c53b84067a0181f5a14e80c13d5acb8150b659f4d73
+size 15984

checkpoint-1650/rng_state_2.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c9fbc9fa428939be10b46779f0eb5cd833e0da426b1cbdee77b3a55b6952235b
+size 15984

checkpoint-1650/rng_state_3.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ac55dba0b79d5fa4699d239da2f966d52040d576d31234ac8d4632e6956481bc
+size 15984

checkpoint-1650/rng_state_4.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:af2d0c015100768ffa23faf3b6c2d54ea89eb045603e30e55cd211e06ff34972
+size 15984

checkpoint-1650/rng_state_5.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c60a1b40608e34bc801c8231f97b81c53b5290dfaed1b9cd0ccbeca29574a991
+size 15984

checkpoint-1650/rng_state_6.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3ad6a142a403eb9aafc4a3a9a856bca648fe31fd22d796867baca31fb13656aa
+size 15984

checkpoint-1650/rng_state_7.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:38bc23a138cc800b22881742c0f3f9a71731a9a7111c6058a0077e6274d21773
+size 15984

checkpoint-1650/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:eca48213d2534f9efd883918137c187e5228038fab577926bdba008860345829
+size 1064

checkpoint-1650/special_tokens_map.json

@@ -0,0 +1,25 @@
+{
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|object_ref_start|>",
+    "<|object_ref_end|>",
+    "<|box_start|>",
+    "<|box_end|>",
+    "<|quad_start|>",
+    "<|quad_end|>",
+    "<|vision_start|>",
+    "<|vision_end|>",
+    "<|vision_pad|>",
+    "<|image_pad|>",
+    "<|video_pad|>"
+  ],
+  "eos_token": {
+    "content": "<|im_end|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "<|im_end|>"
+}

checkpoint-1650/tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9c5ae00e602b8860cbd784ba82a8aa14e8feecec692e7076590d014d7b7fdafa
+size 11421896

checkpoint-1650/tokenizer_config.json

@@ -0,0 +1,208 @@
+{
+  "add_bos_token": false,
+  "add_prefix_space": false,
+  "added_tokens_decoder": {
+    "151643": {
+      "content": "<|endoftext|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151644": {
+      "content": "<|im_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151645": {
+      "content": "<|im_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151646": {
+      "content": "<|object_ref_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151647": {
+      "content": "<|object_ref_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151648": {
+      "content": "<|box_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151649": {
+      "content": "<|box_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151650": {
+      "content": "<|quad_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151651": {
+      "content": "<|quad_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151652": {
+      "content": "<|vision_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151653": {
+      "content": "<|vision_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151654": {
+      "content": "<|vision_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151655": {
+      "content": "<|image_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151656": {
+      "content": "<|video_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151657": {
+      "content": "<tool_call>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151658": {
+      "content": "</tool_call>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151659": {
+      "content": "<|fim_prefix|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151660": {
+      "content": "<|fim_middle|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151661": {
+      "content": "<|fim_suffix|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151662": {
+      "content": "<|fim_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151663": {
+      "content": "<|repo_name|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151664": {
+      "content": "<|file_sep|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    }
+  },
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|object_ref_start|>",
+    "<|object_ref_end|>",
+    "<|box_start|>",
+    "<|box_end|>",
+    "<|quad_start|>",
+    "<|quad_end|>",
+    "<|vision_start|>",
+    "<|vision_end|>",
+    "<|vision_pad|>",
+    "<|image_pad|>",
+    "<|video_pad|>"
+  ],
+  "bos_token": null,
+  "chat_template": "{%- if tools %}\n    {{- '<|im_start|>system\\n' }}\n    {%- if messages[0]['role'] == 'system' %}\n        {{- messages[0]['content'] }}\n    {%- else %}\n        {{- 'Please reason step by step, and put your final answer within \\\\boxed{}.' }}\n    {%- endif %}\n    {{- \"\\n\\n# Tools\\n\\nYou may call one or more functions to assist with the user query.\\n\\nYou are provided with function signatures within <tools></tools> XML tags:\\n<tools>\" }}\n    {%- for tool in tools %}\n        {{- \"\\n\" }}\n        {{- tool | tojson }}\n    {%- endfor %}\n    {{- \"\\n</tools>\\n\\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\\n<tool_call>\\n{\\\"name\\\": <function-name>, \\\"arguments\\\": <args-json-object>}\\n</tool_call><|im_end|>\\n\" }}\n{%- else %}\n    {%- if messages[0]['role'] == 'system' %}\n        {{- '<|im_start|>system\\n' + messages[0]['content'] + '<|im_end|>\\n' }}\n    {%- else %}\n        {{- '<|im_start|>system\\nPlease reason step by step, and put your final answer within \\\\boxed{}.<|im_end|>\\n' }}\n    {%- endif %}\n{%- endif %}\n{%- for message in messages %}\n    {%- if (message.role == \"user\") or (message.role == \"system\" and not loop.first) or (message.role == \"assistant\" and not message.tool_calls) %}\n        {{- '<|im_start|>' + message.role + '\\n' + message.content + '<|im_end|>' + '\\n' }}\n    {%- elif message.role == \"assistant\" %}\n        {{- '<|im_start|>' + message.role }}\n        {%- if message.content %}\n            {{- '\\n' + message.content }}\n        {%- endif %}\n        {%- for tool_call in message.tool_calls %}\n            {%- if tool_call.function is defined %}\n                {%- set tool_call = tool_call.function %}\n            {%- endif %}\n            {{- '\\n<tool_call>\\n{\"name\": \"' }}\n            {{- tool_call.name }}\n            {{- '\", \"arguments\": ' }}\n            {{- tool_call.arguments | tojson }}\n            {{- '}\\n</tool_call>' }}\n        {%- endfor %}\n        {{- '<|im_end|>\\n' }}\n    {%- elif message.role == \"tool\" %}\n        {%- if (loop.index0 == 0) or (messages[loop.index0 - 1].role != \"tool\") %}\n            {{- '<|im_start|>user' }}\n        {%- endif %}\n        {{- '\\n<tool_response>\\n' }}\n        {{- message.content }}\n        {{- '\\n</tool_response>' }}\n        {%- if loop.last or (messages[loop.index0 + 1].role != \"tool\") %}\n            {{- '<|im_end|>\\n' }}\n        {%- endif %}\n    {%- endif %}\n{%- endfor %}\n{%- if add_generation_prompt %}\n    {{- '<|im_start|>assistant\\n' }}\n{%- endif %}\n",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|im_end|>",
+  "errors": "replace",
+  "extra_special_tokens": {},
+  "model_max_length": 131072,
+  "pad_token": "<|im_end|>",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null
+}

checkpoint-1650/trainer_state.json

@@ -0,0 +1,2343 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 1.537744641192917,
+  "eval_steps": 500,
+  "global_step": 1650,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.004659832246039142,
+      "grad_norm": 59.78214918326703,
+      "learning_rate": 7.763975155279503e-07,
+      "loss": 10.6865,
+      "step": 5
+    },
+    {
+      "epoch": 0.009319664492078284,
+      "grad_norm": 61.61821429517621,
+      "learning_rate": 1.5527950310559006e-06,
+      "loss": 10.4783,
+      "step": 10
+    },
+    {
+      "epoch": 0.013979496738117428,
+      "grad_norm": 100.18886340338842,
+      "learning_rate": 2.329192546583851e-06,
+      "loss": 8.8595,
+      "step": 15
+    },
+    {
+      "epoch": 0.01863932898415657,
+      "grad_norm": 28.80282471816165,
+      "learning_rate": 3.1055900621118013e-06,
+      "loss": 2.9699,
+      "step": 20
+    },
+    {
+      "epoch": 0.023299161230195712,
+      "grad_norm": 2.9469993742141027,
+      "learning_rate": 3.881987577639752e-06,
+      "loss": 1.3152,
+      "step": 25
+    },
+    {
+      "epoch": 0.027958993476234855,
+      "grad_norm": 1.273876748177396,
+      "learning_rate": 4.658385093167702e-06,
+      "loss": 0.9979,
+      "step": 30
+    },
+    {
+      "epoch": 0.032618825722273995,
+      "grad_norm": 1.1847847745298432,
+      "learning_rate": 5.4347826086956525e-06,
+      "loss": 0.8462,
+      "step": 35
+    },
+    {
+      "epoch": 0.03727865796831314,
+      "grad_norm": 0.6184161495875878,
+      "learning_rate": 6.2111801242236025e-06,
+      "loss": 0.7823,
+      "step": 40
+    },
+    {
+      "epoch": 0.04193849021435228,
+      "grad_norm": 0.4611063105597401,
+      "learning_rate": 6.9875776397515525e-06,
+      "loss": 0.7195,
+      "step": 45
+    },
+    {
+      "epoch": 0.046598322460391424,
+      "grad_norm": 0.3903179820498685,
+      "learning_rate": 7.763975155279503e-06,
+      "loss": 0.6868,
+      "step": 50
+    },
+    {
+      "epoch": 0.05125815470643057,
+      "grad_norm": 0.44815192407139737,
+      "learning_rate": 8.540372670807453e-06,
+      "loss": 0.6483,
+      "step": 55
+    },
+    {
+      "epoch": 0.05591798695246971,
+      "grad_norm": 0.3363352375310443,
+      "learning_rate": 9.316770186335403e-06,
+      "loss": 0.61,
+      "step": 60
+    },
+    {
+      "epoch": 0.06057781919850885,
+      "grad_norm": 0.31214409078055283,
+      "learning_rate": 1.0093167701863353e-05,
+      "loss": 0.5932,
+      "step": 65
+    },
+    {
+      "epoch": 0.06523765144454799,
+      "grad_norm": 0.32957450493165075,
+      "learning_rate": 1.0869565217391305e-05,
+      "loss": 0.5886,
+      "step": 70
+    },
+    {
+      "epoch": 0.06989748369058714,
+      "grad_norm": 0.3311294918877853,
+      "learning_rate": 1.1645962732919255e-05,
+      "loss": 0.5704,
+      "step": 75
+    },
+    {
+      "epoch": 0.07455731593662628,
+      "grad_norm": 0.3181163049876225,
+      "learning_rate": 1.2422360248447205e-05,
+      "loss": 0.5604,
+      "step": 80
+    },
+    {
+      "epoch": 0.07921714818266543,
+      "grad_norm": 0.3310250029185392,
+      "learning_rate": 1.3198757763975155e-05,
+      "loss": 0.5508,
+      "step": 85
+    },
+    {
+      "epoch": 0.08387698042870456,
+      "grad_norm": 0.2864578039803888,
+      "learning_rate": 1.3975155279503105e-05,
+      "loss": 0.5606,
+      "step": 90
+    },
+    {
+      "epoch": 0.08853681267474371,
+      "grad_norm": 0.2527763382663063,
+      "learning_rate": 1.4751552795031057e-05,
+      "loss": 0.542,
+      "step": 95
+    },
+    {
+      "epoch": 0.09319664492078285,
+      "grad_norm": 0.30977172064299785,
+      "learning_rate": 1.5527950310559007e-05,
+      "loss": 0.5272,
+      "step": 100
+    },
+    {
+      "epoch": 0.097856477166822,
+      "grad_norm": 0.3503262906800675,
+      "learning_rate": 1.630434782608696e-05,
+      "loss": 0.5214,
+      "step": 105
+    },
+    {
+      "epoch": 0.10251630941286113,
+      "grad_norm": 0.3032318880335728,
+      "learning_rate": 1.7080745341614907e-05,
+      "loss": 0.5229,
+      "step": 110
+    },
+    {
+      "epoch": 0.10717614165890028,
+      "grad_norm": 0.32192998759757896,
+      "learning_rate": 1.785714285714286e-05,
+      "loss": 0.5201,
+      "step": 115
+    },
+    {
+      "epoch": 0.11183597390493942,
+      "grad_norm": 0.37017049044979194,
+      "learning_rate": 1.8633540372670807e-05,
+      "loss": 0.5067,
+      "step": 120
+    },
+    {
+      "epoch": 0.11649580615097857,
+      "grad_norm": 0.27625624649694025,
+      "learning_rate": 1.940993788819876e-05,
+      "loss": 0.5026,
+      "step": 125
+    },
+    {
+      "epoch": 0.1211556383970177,
+      "grad_norm": 0.32732522406326287,
+      "learning_rate": 2.0186335403726707e-05,
+      "loss": 0.5083,
+      "step": 130
+    },
+    {
+      "epoch": 0.12581547064305684,
+      "grad_norm": 0.3974302327759709,
+      "learning_rate": 2.096273291925466e-05,
+      "loss": 0.5069,
+      "step": 135
+    },
+    {
+      "epoch": 0.13047530288909598,
+      "grad_norm": 0.49055099062465235,
+      "learning_rate": 2.173913043478261e-05,
+      "loss": 0.4918,
+      "step": 140
+    },
+    {
+      "epoch": 0.13513513513513514,
+      "grad_norm": 0.3509510737287038,
+      "learning_rate": 2.2515527950310562e-05,
+      "loss": 0.5182,
+      "step": 145
+    },
+    {
+      "epoch": 0.13979496738117428,
+      "grad_norm": 0.4060738145091598,
+      "learning_rate": 2.329192546583851e-05,
+      "loss": 0.4924,
+      "step": 150
+    },
+    {
+      "epoch": 0.14445479962721341,
+      "grad_norm": 0.42238178931670933,
+      "learning_rate": 2.4068322981366462e-05,
+      "loss": 0.5005,
+      "step": 155
+    },
+    {
+      "epoch": 0.14911463187325255,
+      "grad_norm": 0.42361270461040995,
+      "learning_rate": 2.484472049689441e-05,
+      "loss": 0.4809,
+      "step": 160
+    },
+    {
+      "epoch": 0.15377446411929171,
+      "grad_norm": 0.4419148082648927,
+      "learning_rate": 2.5621118012422362e-05,
+      "loss": 0.4922,
+      "step": 165
+    },
+    {
+      "epoch": 0.15843429636533085,
+      "grad_norm": 0.37817457175825797,
+      "learning_rate": 2.639751552795031e-05,
+      "loss": 0.4682,
+      "step": 170
+    },
+    {
+      "epoch": 0.16309412861137,
+      "grad_norm": 0.4612740179437555,
+      "learning_rate": 2.7173913043478262e-05,
+      "loss": 0.4812,
+      "step": 175
+    },
+    {
+      "epoch": 0.16775396085740912,
+      "grad_norm": 0.4027204736632852,
+      "learning_rate": 2.795031055900621e-05,
+      "loss": 0.4743,
+      "step": 180
+    },
+    {
+      "epoch": 0.1724137931034483,
+      "grad_norm": 0.3662916369622068,
+      "learning_rate": 2.8726708074534165e-05,
+      "loss": 0.4771,
+      "step": 185
+    },
+    {
+      "epoch": 0.17707362534948742,
+      "grad_norm": 0.44549022951820444,
+      "learning_rate": 2.9503105590062114e-05,
+      "loss": 0.4872,
+      "step": 190
+    },
+    {
+      "epoch": 0.18173345759552656,
+      "grad_norm": 0.4421692278535386,
+      "learning_rate": 3.0279503105590062e-05,
+      "loss": 0.4768,
+      "step": 195
+    },
+    {
+      "epoch": 0.1863932898415657,
+      "grad_norm": 0.4592171701659634,
+      "learning_rate": 3.1055900621118014e-05,
+      "loss": 0.4782,
+      "step": 200
+    },
+    {
+      "epoch": 0.19105312208760486,
+      "grad_norm": 0.5338610618981041,
+      "learning_rate": 3.183229813664597e-05,
+      "loss": 0.4677,
+      "step": 205
+    },
+    {
+      "epoch": 0.195712954333644,
+      "grad_norm": 0.679375515050287,
+      "learning_rate": 3.260869565217392e-05,
+      "loss": 0.4817,
+      "step": 210
+    },
+    {
+      "epoch": 0.20037278657968313,
+      "grad_norm": 0.5075771202954662,
+      "learning_rate": 3.3385093167701865e-05,
+      "loss": 0.4632,
+      "step": 215
+    },
+    {
+      "epoch": 0.20503261882572227,
+      "grad_norm": 0.5271972882853628,
+      "learning_rate": 3.4161490683229814e-05,
+      "loss": 0.4674,
+      "step": 220
+    },
+    {
+      "epoch": 0.2096924510717614,
+      "grad_norm": 0.45927485782401883,
+      "learning_rate": 3.493788819875777e-05,
+      "loss": 0.4496,
+      "step": 225
+    },
+    {
+      "epoch": 0.21435228331780057,
+      "grad_norm": 0.3875430276374643,
+      "learning_rate": 3.571428571428572e-05,
+      "loss": 0.4628,
+      "step": 230
+    },
+    {
+      "epoch": 0.2190121155638397,
+      "grad_norm": 0.43592470909651004,
+      "learning_rate": 3.6490683229813665e-05,
+      "loss": 0.4604,
+      "step": 235
+    },
+    {
+      "epoch": 0.22367194780987884,
+      "grad_norm": 0.45541861707423287,
+      "learning_rate": 3.7267080745341614e-05,
+      "loss": 0.4578,
+      "step": 240
+    },
+    {
+      "epoch": 0.22833178005591798,
+      "grad_norm": 0.5415802082513514,
+      "learning_rate": 3.804347826086957e-05,
+      "loss": 0.4628,
+      "step": 245
+    },
+    {
+      "epoch": 0.23299161230195714,
+      "grad_norm": 0.48756789908879733,
+      "learning_rate": 3.881987577639752e-05,
+      "loss": 0.4551,
+      "step": 250
+    },
+    {
+      "epoch": 0.23765144454799628,
+      "grad_norm": 0.48641029082339876,
+      "learning_rate": 3.9596273291925465e-05,
+      "loss": 0.4636,
+      "step": 255
+    },
+    {
+      "epoch": 0.2423112767940354,
+      "grad_norm": 0.44167537560377024,
+      "learning_rate": 4.0372670807453414e-05,
+      "loss": 0.4584,
+      "step": 260
+    },
+    {
+      "epoch": 0.24697110904007455,
+      "grad_norm": 0.49523500628083084,
+      "learning_rate": 4.114906832298137e-05,
+      "loss": 0.457,
+      "step": 265
+    },
+    {
+      "epoch": 0.2516309412861137,
+      "grad_norm": 0.46362590765498735,
+      "learning_rate": 4.192546583850932e-05,
+      "loss": 0.4553,
+      "step": 270
+    },
+    {
+      "epoch": 0.25629077353215285,
+      "grad_norm": 0.41387594745783063,
+      "learning_rate": 4.270186335403727e-05,
+      "loss": 0.4606,
+      "step": 275
+    },
+    {
+      "epoch": 0.26095060577819196,
+      "grad_norm": 0.4327868844195844,
+      "learning_rate": 4.347826086956522e-05,
+      "loss": 0.4529,
+      "step": 280
+    },
+    {
+      "epoch": 0.2656104380242311,
+      "grad_norm": 0.425878793618421,
+      "learning_rate": 4.425465838509317e-05,
+      "loss": 0.457,
+      "step": 285
+    },
+    {
+      "epoch": 0.2702702702702703,
+      "grad_norm": 0.49167640847055744,
+      "learning_rate": 4.5031055900621124e-05,
+      "loss": 0.4586,
+      "step": 290
+    },
+    {
+      "epoch": 0.2749301025163094,
+      "grad_norm": 0.6223762427590684,
+      "learning_rate": 4.580745341614907e-05,
+      "loss": 0.4627,
+      "step": 295
+    },
+    {
+      "epoch": 0.27958993476234856,
+      "grad_norm": 0.6138447122463503,
+      "learning_rate": 4.658385093167702e-05,
+      "loss": 0.4643,
+      "step": 300
+    },
+    {
+      "epoch": 0.2842497670083877,
+      "grad_norm": 0.5410077879825527,
+      "learning_rate": 4.736024844720497e-05,
+      "loss": 0.4531,
+      "step": 305
+    },
+    {
+      "epoch": 0.28890959925442683,
+      "grad_norm": 0.5248106064020833,
+      "learning_rate": 4.8136645962732924e-05,
+      "loss": 0.4453,
+      "step": 310
+    },
+    {
+      "epoch": 0.293569431500466,
+      "grad_norm": 0.4576312238038398,
+      "learning_rate": 4.891304347826087e-05,
+      "loss": 0.4531,
+      "step": 315
+    },
+    {
+      "epoch": 0.2982292637465051,
+      "grad_norm": 0.7616375740291316,
+      "learning_rate": 4.968944099378882e-05,
+      "loss": 0.4387,
+      "step": 320
+    },
+    {
+      "epoch": 0.30288909599254427,
+      "grad_norm": 0.5230246761762287,
+      "learning_rate": 4.994822229892993e-05,
+      "loss": 0.4367,
+      "step": 325
+    },
+    {
+      "epoch": 0.30754892823858343,
+      "grad_norm": 0.4333888474147187,
+      "learning_rate": 4.986192613047981e-05,
+      "loss": 0.4468,
+      "step": 330
+    },
+    {
+      "epoch": 0.31220876048462254,
+      "grad_norm": 0.6558069118015527,
+      "learning_rate": 4.977562996202969e-05,
+      "loss": 0.447,
+      "step": 335
+    },
+    {
+      "epoch": 0.3168685927306617,
+      "grad_norm": 0.4781469048268466,
+      "learning_rate": 4.968933379357957e-05,
+      "loss": 0.4487,
+      "step": 340
+    },
+    {
+      "epoch": 0.32152842497670087,
+      "grad_norm": 0.4582486158349481,
+      "learning_rate": 4.9603037625129445e-05,
+      "loss": 0.4433,
+      "step": 345
+    },
+    {
+      "epoch": 0.32618825722274,
+      "grad_norm": 0.4234664780262906,
+      "learning_rate": 4.951674145667933e-05,
+      "loss": 0.4568,
+      "step": 350
+    },
+    {
+      "epoch": 0.33084808946877914,
+      "grad_norm": 0.4766441062632173,
+      "learning_rate": 4.94304452882292e-05,
+      "loss": 0.4431,
+      "step": 355
+    },
+    {
+      "epoch": 0.33550792171481825,
+      "grad_norm": 0.5294140666427822,
+      "learning_rate": 4.934414911977908e-05,
+      "loss": 0.4405,
+      "step": 360
+    },
+    {
+      "epoch": 0.3401677539608574,
+      "grad_norm": 0.3838225157756671,
+      "learning_rate": 4.9257852951328965e-05,
+      "loss": 0.437,
+      "step": 365
+    },
+    {
+      "epoch": 0.3448275862068966,
+      "grad_norm": 0.4289011151413574,
+      "learning_rate": 4.917155678287884e-05,
+      "loss": 0.4498,
+      "step": 370
+    },
+    {
+      "epoch": 0.3494874184529357,
+      "grad_norm": 0.4739139367556648,
+      "learning_rate": 4.908526061442872e-05,
+      "loss": 0.4364,
+      "step": 375
+    },
+    {
+      "epoch": 0.35414725069897485,
+      "grad_norm": 0.6531685734098429,
+      "learning_rate": 4.89989644459786e-05,
+      "loss": 0.4507,
+      "step": 380
+    },
+    {
+      "epoch": 0.35880708294501396,
+      "grad_norm": 0.38736077721702633,
+      "learning_rate": 4.891266827752848e-05,
+      "loss": 0.4526,
+      "step": 385
+    },
+    {
+      "epoch": 0.3634669151910531,
+      "grad_norm": 0.39074723361967706,
+      "learning_rate": 4.882637210907836e-05,
+      "loss": 0.4372,
+      "step": 390
+    },
+    {
+      "epoch": 0.3681267474370923,
+      "grad_norm": 0.4998720672786241,
+      "learning_rate": 4.874007594062824e-05,
+      "loss": 0.4432,
+      "step": 395
+    },
+    {
+      "epoch": 0.3727865796831314,
+      "grad_norm": 0.41449696680501236,
+      "learning_rate": 4.865377977217811e-05,
+      "loss": 0.4284,
+      "step": 400
+    },
+    {
+      "epoch": 0.37744641192917056,
+      "grad_norm": 0.42798981510455325,
+      "learning_rate": 4.8567483603728e-05,
+      "loss": 0.4466,
+      "step": 405
+    },
+    {
+      "epoch": 0.3821062441752097,
+      "grad_norm": 0.453510571603012,
+      "learning_rate": 4.8481187435277875e-05,
+      "loss": 0.4425,
+      "step": 410
+    },
+    {
+      "epoch": 0.38676607642124883,
+      "grad_norm": 0.5613852564226437,
+      "learning_rate": 4.839489126682776e-05,
+      "loss": 0.4296,
+      "step": 415
+    },
+    {
+      "epoch": 0.391425908667288,
+      "grad_norm": 0.4991311823287778,
+      "learning_rate": 4.830859509837763e-05,
+      "loss": 0.4477,
+      "step": 420
+    },
+    {
+      "epoch": 0.3960857409133271,
+      "grad_norm": 0.41710093216407557,
+      "learning_rate": 4.822229892992751e-05,
+      "loss": 0.4451,
+      "step": 425
+    },
+    {
+      "epoch": 0.40074557315936626,
+      "grad_norm": 0.44852195468081424,
+      "learning_rate": 4.8136002761477395e-05,
+      "loss": 0.4322,
+      "step": 430
+    },
+    {
+      "epoch": 0.40540540540540543,
+      "grad_norm": 0.5873258297489329,
+      "learning_rate": 4.804970659302727e-05,
+      "loss": 0.445,
+      "step": 435
+    },
+    {
+      "epoch": 0.41006523765144454,
+      "grad_norm": 0.5301184440251494,
+      "learning_rate": 4.796341042457715e-05,
+      "loss": 0.439,
+      "step": 440
+    },
+    {
+      "epoch": 0.4147250698974837,
+      "grad_norm": 0.554152067322795,
+      "learning_rate": 4.787711425612703e-05,
+      "loss": 0.4337,
+      "step": 445
+    },
+    {
+      "epoch": 0.4193849021435228,
+      "grad_norm": 0.4875794890348032,
+      "learning_rate": 4.779081808767691e-05,
+      "loss": 0.4245,
+      "step": 450
+    },
+    {
+      "epoch": 0.424044734389562,
+      "grad_norm": 0.4690783572871423,
+      "learning_rate": 4.770452191922679e-05,
+      "loss": 0.4314,
+      "step": 455
+    },
+    {
+      "epoch": 0.42870456663560114,
+      "grad_norm": 0.3725289513240759,
+      "learning_rate": 4.761822575077667e-05,
+      "loss": 0.4283,
+      "step": 460
+    },
+    {
+      "epoch": 0.43336439888164024,
+      "grad_norm": 0.4830268598668616,
+      "learning_rate": 4.753192958232654e-05,
+      "loss": 0.4255,
+      "step": 465
+    },
+    {
+      "epoch": 0.4380242311276794,
+      "grad_norm": 0.43173494250112954,
+      "learning_rate": 4.744563341387643e-05,
+      "loss": 0.4378,
+      "step": 470
+    },
+    {
+      "epoch": 0.4426840633737186,
+      "grad_norm": 0.43237002431737065,
+      "learning_rate": 4.7359337245426306e-05,
+      "loss": 0.4277,
+      "step": 475
+    },
+    {
+      "epoch": 0.4473438956197577,
+      "grad_norm": 0.41385681702794824,
+      "learning_rate": 4.7273041076976184e-05,
+      "loss": 0.4394,
+      "step": 480
+    },
+    {
+      "epoch": 0.45200372786579684,
+      "grad_norm": 0.40157124060011173,
+      "learning_rate": 4.718674490852606e-05,
+      "loss": 0.432,
+      "step": 485
+    },
+    {
+      "epoch": 0.45666356011183595,
+      "grad_norm": 0.39938983254093463,
+      "learning_rate": 4.710044874007594e-05,
+      "loss": 0.4264,
+      "step": 490
+    },
+    {
+      "epoch": 0.4613233923578751,
+      "grad_norm": 0.39732323279012777,
+      "learning_rate": 4.7014152571625826e-05,
+      "loss": 0.4321,
+      "step": 495
+    },
+    {
+      "epoch": 0.4659832246039143,
+      "grad_norm": 0.4747358464791143,
+      "learning_rate": 4.6927856403175704e-05,
+      "loss": 0.435,
+      "step": 500
+    },
+    {
+      "epoch": 0.4706430568499534,
+      "grad_norm": 0.3698718174123855,
+      "learning_rate": 4.684156023472558e-05,
+      "loss": 0.4221,
+      "step": 505
+    },
+    {
+      "epoch": 0.47530288909599255,
+      "grad_norm": 0.4305572996344627,
+      "learning_rate": 4.675526406627546e-05,
+      "loss": 0.4303,
+      "step": 510
+    },
+    {
+      "epoch": 0.47996272134203166,
+      "grad_norm": 0.6085259797324423,
+      "learning_rate": 4.666896789782534e-05,
+      "loss": 0.4281,
+      "step": 515
+    },
+    {
+      "epoch": 0.4846225535880708,
+      "grad_norm": 0.5730318489213171,
+      "learning_rate": 4.658267172937522e-05,
+      "loss": 0.4321,
+      "step": 520
+    },
+    {
+      "epoch": 0.48928238583411,
+      "grad_norm": 0.4332245949035479,
+      "learning_rate": 4.64963755609251e-05,
+      "loss": 0.4309,
+      "step": 525
+    },
+    {
+      "epoch": 0.4939422180801491,
+      "grad_norm": 0.508102013567185,
+      "learning_rate": 4.641007939247497e-05,
+      "loss": 0.428,
+      "step": 530
+    },
+    {
+      "epoch": 0.49860205032618826,
+      "grad_norm": 0.34669842614662666,
+      "learning_rate": 4.632378322402486e-05,
+      "loss": 0.4283,
+      "step": 535
+    },
+    {
+      "epoch": 0.5032618825722274,
+      "grad_norm": 0.3889254150420956,
+      "learning_rate": 4.6237487055574736e-05,
+      "loss": 0.4178,
+      "step": 540
+    },
+    {
+      "epoch": 0.5079217148182665,
+      "grad_norm": 0.49239466237923585,
+      "learning_rate": 4.6151190887124615e-05,
+      "loss": 0.4244,
+      "step": 545
+    },
+    {
+      "epoch": 0.5125815470643057,
+      "grad_norm": 0.4397316581317278,
+      "learning_rate": 4.606489471867449e-05,
+      "loss": 0.4245,
+      "step": 550
+    },
+    {
+      "epoch": 0.5172413793103449,
+      "grad_norm": 0.3573410355093717,
+      "learning_rate": 4.597859855022437e-05,
+      "loss": 0.4192,
+      "step": 555
+    },
+    {
+      "epoch": 0.5219012115563839,
+      "grad_norm": 0.43267928489519075,
+      "learning_rate": 4.589230238177425e-05,
+      "loss": 0.4397,
+      "step": 560
+    },
+    {
+      "epoch": 0.5265610438024231,
+      "grad_norm": 0.4821719775119577,
+      "learning_rate": 4.5806006213324134e-05,
+      "loss": 0.4177,
+      "step": 565
+    },
+    {
+      "epoch": 0.5312208760484622,
+      "grad_norm": 0.5349367976109402,
+      "learning_rate": 4.5719710044874006e-05,
+      "loss": 0.4175,
+      "step": 570
+    },
+    {
+      "epoch": 0.5358807082945014,
+      "grad_norm": 0.47146780494171403,
+      "learning_rate": 4.563341387642389e-05,
+      "loss": 0.4234,
+      "step": 575
+    },
+    {
+      "epoch": 0.5405405405405406,
+      "grad_norm": 0.3724800991591985,
+      "learning_rate": 4.554711770797377e-05,
+      "loss": 0.4238,
+      "step": 580
+    },
+    {
+      "epoch": 0.5452003727865797,
+      "grad_norm": 0.46072368721402923,
+      "learning_rate": 4.546082153952365e-05,
+      "loss": 0.4209,
+      "step": 585
+    },
+    {
+      "epoch": 0.5498602050326188,
+      "grad_norm": 0.40398358486961483,
+      "learning_rate": 4.5374525371073526e-05,
+      "loss": 0.4171,
+      "step": 590
+    },
+    {
+      "epoch": 0.554520037278658,
+      "grad_norm": 0.3510727726291796,
+      "learning_rate": 4.5288229202623404e-05,
+      "loss": 0.4191,
+      "step": 595
+    },
+    {
+      "epoch": 0.5591798695246971,
+      "grad_norm": 0.38603847092047144,
+      "learning_rate": 4.520193303417328e-05,
+      "loss": 0.4194,
+      "step": 600
+    },
+    {
+      "epoch": 0.5638397017707363,
+      "grad_norm": 0.3689646251475895,
+      "learning_rate": 4.511563686572317e-05,
+      "loss": 0.4182,
+      "step": 605
+    },
+    {
+      "epoch": 0.5684995340167754,
+      "grad_norm": 0.408796996489251,
+      "learning_rate": 4.5029340697273045e-05,
+      "loss": 0.4293,
+      "step": 610
+    },
+    {
+      "epoch": 0.5731593662628145,
+      "grad_norm": 0.32130017833210983,
+      "learning_rate": 4.4943044528822923e-05,
+      "loss": 0.4159,
+      "step": 615
+    },
+    {
+      "epoch": 0.5778191985088537,
+      "grad_norm": 0.3772249822258766,
+      "learning_rate": 4.48567483603728e-05,
+      "loss": 0.4131,
+      "step": 620
+    },
+    {
+      "epoch": 0.5824790307548928,
+      "grad_norm": 0.35575898721323274,
+      "learning_rate": 4.477045219192268e-05,
+      "loss": 0.4344,
+      "step": 625
+    },
+    {
+      "epoch": 0.587138863000932,
+      "grad_norm": 0.38751558564769756,
+      "learning_rate": 4.4684156023472565e-05,
+      "loss": 0.425,
+      "step": 630
+    },
+    {
+      "epoch": 0.5917986952469712,
+      "grad_norm": 0.4663549933535284,
+      "learning_rate": 4.4597859855022436e-05,
+      "loss": 0.4296,
+      "step": 635
+    },
+    {
+      "epoch": 0.5964585274930102,
+      "grad_norm": 0.4662472899566146,
+      "learning_rate": 4.4511563686572315e-05,
+      "loss": 0.4202,
+      "step": 640
+    },
+    {
+      "epoch": 0.6011183597390494,
+      "grad_norm": 0.39559722373002326,
+      "learning_rate": 4.44252675181222e-05,
+      "loss": 0.4197,
+      "step": 645
+    },
+    {
+      "epoch": 0.6057781919850885,
+      "grad_norm": 0.4348073933035661,
+      "learning_rate": 4.433897134967208e-05,
+      "loss": 0.4284,
+      "step": 650
+    },
+    {
+      "epoch": 0.6104380242311277,
+      "grad_norm": 0.4064242144273277,
+      "learning_rate": 4.4252675181221956e-05,
+      "loss": 0.4262,
+      "step": 655
+    },
+    {
+      "epoch": 0.6150978564771669,
+      "grad_norm": 0.3413504824042106,
+      "learning_rate": 4.4166379012771834e-05,
+      "loss": 0.408,
+      "step": 660
+    },
+    {
+      "epoch": 0.6197576887232059,
+      "grad_norm": 0.4625137937805045,
+      "learning_rate": 4.408008284432171e-05,
+      "loss": 0.4091,
+      "step": 665
+    },
+    {
+      "epoch": 0.6244175209692451,
+      "grad_norm": 0.4365632325921953,
+      "learning_rate": 4.39937866758716e-05,
+      "loss": 0.4076,
+      "step": 670
+    },
+    {
+      "epoch": 0.6290773532152842,
+      "grad_norm": 0.43623665649462795,
+      "learning_rate": 4.3907490507421476e-05,
+      "loss": 0.4158,
+      "step": 675
+    },
+    {
+      "epoch": 0.6337371854613234,
+      "grad_norm": 0.4462282373794903,
+      "learning_rate": 4.382119433897135e-05,
+      "loss": 0.419,
+      "step": 680
+    },
+    {
+      "epoch": 0.6383970177073626,
+      "grad_norm": 0.4909062825850483,
+      "learning_rate": 4.373489817052123e-05,
+      "loss": 0.4284,
+      "step": 685
+    },
+    {
+      "epoch": 0.6430568499534017,
+      "grad_norm": 0.4261067060277007,
+      "learning_rate": 4.364860200207111e-05,
+      "loss": 0.4108,
+      "step": 690
+    },
+    {
+      "epoch": 0.6477166821994408,
+      "grad_norm": 0.36806589117120897,
+      "learning_rate": 4.356230583362099e-05,
+      "loss": 0.4154,
+      "step": 695
+    },
+    {
+      "epoch": 0.65237651444548,
+      "grad_norm": 0.34856157577107083,
+      "learning_rate": 4.347600966517087e-05,
+      "loss": 0.4246,
+      "step": 700
+    },
+    {
+      "epoch": 0.6570363466915191,
+      "grad_norm": 0.35691903362815613,
+      "learning_rate": 4.3389713496720745e-05,
+      "loss": 0.4178,
+      "step": 705
+    },
+    {
+      "epoch": 0.6616961789375583,
+      "grad_norm": 0.4185903662996916,
+      "learning_rate": 4.330341732827063e-05,
+      "loss": 0.4184,
+      "step": 710
+    },
+    {
+      "epoch": 0.6663560111835974,
+      "grad_norm": 0.36798569357808036,
+      "learning_rate": 4.321712115982051e-05,
+      "loss": 0.4067,
+      "step": 715
+    },
+    {
+      "epoch": 0.6710158434296365,
+      "grad_norm": 0.3788649720605493,
+      "learning_rate": 4.3130824991370387e-05,
+      "loss": 0.417,
+      "step": 720
+    },
+    {
+      "epoch": 0.6756756756756757,
+      "grad_norm": 0.33269308559456673,
+      "learning_rate": 4.3044528822920265e-05,
+      "loss": 0.4158,
+      "step": 725
+    },
+    {
+      "epoch": 0.6803355079217148,
+      "grad_norm": 0.41281885956590714,
+      "learning_rate": 4.295823265447014e-05,
+      "loss": 0.4107,
+      "step": 730
+    },
+    {
+      "epoch": 0.684995340167754,
+      "grad_norm": 0.35535706301183545,
+      "learning_rate": 4.287193648602002e-05,
+      "loss": 0.4149,
+      "step": 735
+    },
+    {
+      "epoch": 0.6896551724137931,
+      "grad_norm": 0.3277576836962773,
+      "learning_rate": 4.27856403175699e-05,
+      "loss": 0.4086,
+      "step": 740
+    },
+    {
+      "epoch": 0.6943150046598322,
+      "grad_norm": 0.40279395134887447,
+      "learning_rate": 4.269934414911978e-05,
+      "loss": 0.4043,
+      "step": 745
+    },
+    {
+      "epoch": 0.6989748369058714,
+      "grad_norm": 0.43373381741082817,
+      "learning_rate": 4.261304798066966e-05,
+      "loss": 0.4089,
+      "step": 750
+    },
+    {
+      "epoch": 0.7036346691519105,
+      "grad_norm": 0.36906806847778895,
+      "learning_rate": 4.252675181221954e-05,
+      "loss": 0.4113,
+      "step": 755
+    },
+    {
+      "epoch": 0.7082945013979497,
+      "grad_norm": 0.41940285749229916,
+      "learning_rate": 4.244045564376942e-05,
+      "loss": 0.4145,
+      "step": 760
+    },
+    {
+      "epoch": 0.7129543336439889,
+      "grad_norm": 0.35669549672804485,
+      "learning_rate": 4.23541594753193e-05,
+      "loss": 0.3988,
+      "step": 765
+    },
+    {
+      "epoch": 0.7176141658900279,
+      "grad_norm": 0.3590318963506859,
+      "learning_rate": 4.2267863306869176e-05,
+      "loss": 0.4075,
+      "step": 770
+    },
+    {
+      "epoch": 0.7222739981360671,
+      "grad_norm": 0.40077547109104333,
+      "learning_rate": 4.2181567138419054e-05,
+      "loss": 0.4202,
+      "step": 775
+    },
+    {
+      "epoch": 0.7269338303821062,
+      "grad_norm": 0.37047082422653643,
+      "learning_rate": 4.209527096996894e-05,
+      "loss": 0.4049,
+      "step": 780
+    },
+    {
+      "epoch": 0.7315936626281454,
+      "grad_norm": 0.3795501401345936,
+      "learning_rate": 4.200897480151881e-05,
+      "loss": 0.4121,
+      "step": 785
+    },
+    {
+      "epoch": 0.7362534948741846,
+      "grad_norm": 0.4503984514039821,
+      "learning_rate": 4.1922678633068695e-05,
+      "loss": 0.4152,
+      "step": 790
+    },
+    {
+      "epoch": 0.7409133271202236,
+      "grad_norm": 0.44284905780951017,
+      "learning_rate": 4.1836382464618573e-05,
+      "loss": 0.4098,
+      "step": 795
+    },
+    {
+      "epoch": 0.7455731593662628,
+      "grad_norm": 0.44691031160954076,
+      "learning_rate": 4.175008629616845e-05,
+      "loss": 0.4078,
+      "step": 800
+    },
+    {
+      "epoch": 0.750232991612302,
+      "grad_norm": 0.3735786310769167,
+      "learning_rate": 4.166379012771833e-05,
+      "loss": 0.4077,
+      "step": 805
+    },
+    {
+      "epoch": 0.7548928238583411,
+      "grad_norm": 0.3384286028508436,
+      "learning_rate": 4.157749395926821e-05,
+      "loss": 0.4257,
+      "step": 810
+    },
+    {
+      "epoch": 0.7595526561043803,
+      "grad_norm": 0.38501244831095327,
+      "learning_rate": 4.1491197790818086e-05,
+      "loss": 0.4005,
+      "step": 815
+    },
+    {
+      "epoch": 0.7642124883504194,
+      "grad_norm": 0.6658921381907594,
+      "learning_rate": 4.140490162236797e-05,
+      "loss": 0.4152,
+      "step": 820
+    },
+    {
+      "epoch": 0.7688723205964585,
+      "grad_norm": 0.5108682931365331,
+      "learning_rate": 4.131860545391785e-05,
+      "loss": 0.4078,
+      "step": 825
+    },
+    {
+      "epoch": 0.7735321528424977,
+      "grad_norm": 0.3959267206623357,
+      "learning_rate": 4.123230928546773e-05,
+      "loss": 0.4025,
+      "step": 830
+    },
+    {
+      "epoch": 0.7781919850885368,
+      "grad_norm": 0.40541297925684416,
+      "learning_rate": 4.1146013117017606e-05,
+      "loss": 0.4095,
+      "step": 835
+    },
+    {
+      "epoch": 0.782851817334576,
+      "grad_norm": 0.4483501398906087,
+      "learning_rate": 4.1059716948567484e-05,
+      "loss": 0.4127,
+      "step": 840
+    },
+    {
+      "epoch": 0.7875116495806151,
+      "grad_norm": 0.4577963072705479,
+      "learning_rate": 4.097342078011737e-05,
+      "loss": 0.4204,
+      "step": 845
+    },
+    {
+      "epoch": 0.7921714818266542,
+      "grad_norm": 0.3657002203284406,
+      "learning_rate": 4.088712461166724e-05,
+      "loss": 0.4105,
+      "step": 850
+    },
+    {
+      "epoch": 0.7968313140726934,
+      "grad_norm": 0.5219753962445839,
+      "learning_rate": 4.080082844321712e-05,
+      "loss": 0.4121,
+      "step": 855
+    },
+    {
+      "epoch": 0.8014911463187325,
+      "grad_norm": 0.4051529596962449,
+      "learning_rate": 4.0714532274767004e-05,
+      "loss": 0.4189,
+      "step": 860
+    },
+    {
+      "epoch": 0.8061509785647717,
+      "grad_norm": 0.3645428936825488,
+      "learning_rate": 4.062823610631688e-05,
+      "loss": 0.4003,
+      "step": 865
+    },
+    {
+      "epoch": 0.8108108108108109,
+      "grad_norm": 0.31752471962451867,
+      "learning_rate": 4.054193993786676e-05,
+      "loss": 0.415,
+      "step": 870
+    },
+    {
+      "epoch": 0.8154706430568499,
+      "grad_norm": 0.3601862957329616,
+      "learning_rate": 4.045564376941664e-05,
+      "loss": 0.4112,
+      "step": 875
+    },
+    {
+      "epoch": 0.8201304753028891,
+      "grad_norm": 0.36233941550256943,
+      "learning_rate": 4.036934760096652e-05,
+      "loss": 0.4027,
+      "step": 880
+    },
+    {
+      "epoch": 0.8247903075489282,
+      "grad_norm": 0.41290807812840935,
+      "learning_rate": 4.02830514325164e-05,
+      "loss": 0.4159,
+      "step": 885
+    },
+    {
+      "epoch": 0.8294501397949674,
+      "grad_norm": 0.42156370600929094,
+      "learning_rate": 4.019675526406628e-05,
+      "loss": 0.4026,
+      "step": 890
+    },
+    {
+      "epoch": 0.8341099720410066,
+      "grad_norm": 0.369144224526896,
+      "learning_rate": 4.011045909561615e-05,
+      "loss": 0.4216,
+      "step": 895
+    },
+    {
+      "epoch": 0.8387698042870456,
+      "grad_norm": 0.35301766518057,
+      "learning_rate": 4.0024162927166037e-05,
+      "loss": 0.4047,
+      "step": 900
+    },
+    {
+      "epoch": 0.8434296365330848,
+      "grad_norm": 0.30727777017232927,
+      "learning_rate": 3.9937866758715915e-05,
+      "loss": 0.4061,
+      "step": 905
+    },
+    {
+      "epoch": 0.848089468779124,
+      "grad_norm": 0.4595118441196378,
+      "learning_rate": 3.98515705902658e-05,
+      "loss": 0.4121,
+      "step": 910
+    },
+    {
+      "epoch": 0.8527493010251631,
+      "grad_norm": 0.5868283893854741,
+      "learning_rate": 3.976527442181567e-05,
+      "loss": 0.4051,
+      "step": 915
+    },
+    {
+      "epoch": 0.8574091332712023,
+      "grad_norm": 0.4408220457405354,
+      "learning_rate": 3.967897825336555e-05,
+      "loss": 0.4113,
+      "step": 920
+    },
+    {
+      "epoch": 0.8620689655172413,
+      "grad_norm": 0.329430445296284,
+      "learning_rate": 3.9592682084915434e-05,
+      "loss": 0.398,
+      "step": 925
+    },
+    {
+      "epoch": 0.8667287977632805,
+      "grad_norm": 0.4772951342481958,
+      "learning_rate": 3.950638591646531e-05,
+      "loss": 0.398,
+      "step": 930
+    },
+    {
+      "epoch": 0.8713886300093197,
+      "grad_norm": 0.39009508489962785,
+      "learning_rate": 3.942008974801519e-05,
+      "loss": 0.4091,
+      "step": 935
+    },
+    {
+      "epoch": 0.8760484622553588,
+      "grad_norm": 0.31283951592729364,
+      "learning_rate": 3.933379357956507e-05,
+      "loss": 0.4051,
+      "step": 940
+    },
+    {
+      "epoch": 0.880708294501398,
+      "grad_norm": 0.38362272885626797,
+      "learning_rate": 3.924749741111495e-05,
+      "loss": 0.4072,
+      "step": 945
+    },
+    {
+      "epoch": 0.8853681267474371,
+      "grad_norm": 0.4294812901025146,
+      "learning_rate": 3.916120124266483e-05,
+      "loss": 0.4148,
+      "step": 950
+    },
+    {
+      "epoch": 0.8900279589934762,
+      "grad_norm": 0.4001848027163216,
+      "learning_rate": 3.9074905074214704e-05,
+      "loss": 0.4136,
+      "step": 955
+    },
+    {
+      "epoch": 0.8946877912395154,
+      "grad_norm": 0.34026388057585993,
+      "learning_rate": 3.898860890576458e-05,
+      "loss": 0.4093,
+      "step": 960
+    },
+    {
+      "epoch": 0.8993476234855545,
+      "grad_norm": 0.40668203338539827,
+      "learning_rate": 3.890231273731447e-05,
+      "loss": 0.4008,
+      "step": 965
+    },
+    {
+      "epoch": 0.9040074557315937,
+      "grad_norm": 0.3294333705662259,
+      "learning_rate": 3.8816016568864345e-05,
+      "loss": 0.3987,
+      "step": 970
+    },
+    {
+      "epoch": 0.9086672879776329,
+      "grad_norm": 0.342795942254618,
+      "learning_rate": 3.8729720400414224e-05,
+      "loss": 0.4176,
+      "step": 975
+    },
+    {
+      "epoch": 0.9133271202236719,
+      "grad_norm": 0.39628105185451024,
+      "learning_rate": 3.86434242319641e-05,
+      "loss": 0.4047,
+      "step": 980
+    },
+    {
+      "epoch": 0.9179869524697111,
+      "grad_norm": 0.46631619804138,
+      "learning_rate": 3.855712806351398e-05,
+      "loss": 0.4068,
+      "step": 985
+    },
+    {
+      "epoch": 0.9226467847157502,
+      "grad_norm": 0.43261058596610696,
+      "learning_rate": 3.8470831895063865e-05,
+      "loss": 0.4038,
+      "step": 990
+    },
+    {
+      "epoch": 0.9273066169617894,
+      "grad_norm": 0.37541980348762116,
+      "learning_rate": 3.838453572661374e-05,
+      "loss": 0.3995,
+      "step": 995
+    },
+    {
+      "epoch": 0.9319664492078286,
+      "grad_norm": 0.4184847694680633,
+      "learning_rate": 3.8298239558163615e-05,
+      "loss": 0.4025,
+      "step": 1000
+    },
+    {
+      "epoch": 0.9366262814538676,
+      "grad_norm": 0.3857591059837171,
+      "learning_rate": 3.82119433897135e-05,
+      "loss": 0.4075,
+      "step": 1005
+    },
+    {
+      "epoch": 0.9412861136999068,
+      "grad_norm": 0.39800461331863696,
+      "learning_rate": 3.812564722126338e-05,
+      "loss": 0.4,
+      "step": 1010
+    },
+    {
+      "epoch": 0.9459459459459459,
+      "grad_norm": 0.4143464836173206,
+      "learning_rate": 3.8039351052813256e-05,
+      "loss": 0.4083,
+      "step": 1015
+    },
+    {
+      "epoch": 0.9506057781919851,
+      "grad_norm": 0.3920202302045243,
+      "learning_rate": 3.7953054884363134e-05,
+      "loss": 0.4023,
+      "step": 1020
+    },
+    {
+      "epoch": 0.9552656104380243,
+      "grad_norm": 0.37828355875661107,
+      "learning_rate": 3.786675871591301e-05,
+      "loss": 0.4056,
+      "step": 1025
+    },
+    {
+      "epoch": 0.9599254426840633,
+      "grad_norm": 0.29419352669408066,
+      "learning_rate": 3.77804625474629e-05,
+      "loss": 0.4107,
+      "step": 1030
+    },
+    {
+      "epoch": 0.9645852749301025,
+      "grad_norm": 0.3997612657539297,
+      "learning_rate": 3.7694166379012776e-05,
+      "loss": 0.4031,
+      "step": 1035
+    },
+    {
+      "epoch": 0.9692451071761417,
+      "grad_norm": 0.38608182016741566,
+      "learning_rate": 3.7607870210562654e-05,
+      "loss": 0.4005,
+      "step": 1040
+    },
+    {
+      "epoch": 0.9739049394221808,
+      "grad_norm": 0.28600031302754375,
+      "learning_rate": 3.752157404211253e-05,
+      "loss": 0.3933,
+      "step": 1045
+    },
+    {
+      "epoch": 0.97856477166822,
+      "grad_norm": 0.38107411335734614,
+      "learning_rate": 3.743527787366241e-05,
+      "loss": 0.4225,
+      "step": 1050
+    },
+    {
+      "epoch": 0.983224603914259,
+      "grad_norm": 0.39170807411989667,
+      "learning_rate": 3.734898170521229e-05,
+      "loss": 0.4004,
+      "step": 1055
+    },
+    {
+      "epoch": 0.9878844361602982,
+      "grad_norm": 0.38467780448535566,
+      "learning_rate": 3.7262685536762174e-05,
+      "loss": 0.3997,
+      "step": 1060
+    },
+    {
+      "epoch": 0.9925442684063374,
+      "grad_norm": 0.4020140968286371,
+      "learning_rate": 3.7176389368312045e-05,
+      "loss": 0.4037,
+      "step": 1065
+    },
+    {
+      "epoch": 0.9972041006523765,
+      "grad_norm": 0.46309836470033766,
+      "learning_rate": 3.709009319986193e-05,
+      "loss": 0.4041,
+      "step": 1070
+    },
+    {
+      "epoch": 1.0018639328984156,
+      "grad_norm": 0.3901703886810142,
+      "learning_rate": 3.700379703141181e-05,
+      "loss": 0.369,
+      "step": 1075
+    },
+    {
+      "epoch": 1.0065237651444547,
+      "grad_norm": 0.3010234352246036,
+      "learning_rate": 3.6917500862961687e-05,
+      "loss": 0.3448,
+      "step": 1080
+    },
+    {
+      "epoch": 1.011183597390494,
+      "grad_norm": 0.4157638937857459,
+      "learning_rate": 3.6831204694511565e-05,
+      "loss": 0.3518,
+      "step": 1085
+    },
+    {
+      "epoch": 1.015843429636533,
+      "grad_norm": 0.3603522336344581,
+      "learning_rate": 3.674490852606144e-05,
+      "loss": 0.3586,
+      "step": 1090
+    },
+    {
+      "epoch": 1.0205032618825722,
+      "grad_norm": 0.32822298799590405,
+      "learning_rate": 3.665861235761132e-05,
+      "loss": 0.3447,
+      "step": 1095
+    },
+    {
+      "epoch": 1.0251630941286114,
+      "grad_norm": 0.35038116243974443,
+      "learning_rate": 3.6572316189161206e-05,
+      "loss": 0.337,
+      "step": 1100
+    },
+    {
+      "epoch": 1.0298229263746506,
+      "grad_norm": 0.2841436076230565,
+      "learning_rate": 3.6486020020711085e-05,
+      "loss": 0.343,
+      "step": 1105
+    },
+    {
+      "epoch": 1.0344827586206897,
+      "grad_norm": 0.3116345416367525,
+      "learning_rate": 3.639972385226096e-05,
+      "loss": 0.3432,
+      "step": 1110
+    },
+    {
+      "epoch": 1.0391425908667289,
+      "grad_norm": 0.3151855230791428,
+      "learning_rate": 3.631342768381084e-05,
+      "loss": 0.3431,
+      "step": 1115
+    },
+    {
+      "epoch": 1.0438024231127678,
+      "grad_norm": 0.35993766510002373,
+      "learning_rate": 3.622713151536072e-05,
+      "loss": 0.343,
+      "step": 1120
+    },
+    {
+      "epoch": 1.048462255358807,
+      "grad_norm": 0.32743934813623815,
+      "learning_rate": 3.6140835346910604e-05,
+      "loss": 0.347,
+      "step": 1125
+    },
+    {
+      "epoch": 1.0531220876048462,
+      "grad_norm": 0.3000370756870265,
+      "learning_rate": 3.6054539178460476e-05,
+      "loss": 0.3433,
+      "step": 1130
+    },
+    {
+      "epoch": 1.0577819198508853,
+      "grad_norm": 0.38781790959885876,
+      "learning_rate": 3.5968243010010354e-05,
+      "loss": 0.3594,
+      "step": 1135
+    },
+    {
+      "epoch": 1.0624417520969245,
+      "grad_norm": 0.32014554787819977,
+      "learning_rate": 3.588194684156024e-05,
+      "loss": 0.3503,
+      "step": 1140
+    },
+    {
+      "epoch": 1.0671015843429636,
+      "grad_norm": 0.370735809498283,
+      "learning_rate": 3.579565067311012e-05,
+      "loss": 0.3417,
+      "step": 1145
+    },
+    {
+      "epoch": 1.0717614165890028,
+      "grad_norm": 0.3024601758163409,
+      "learning_rate": 3.5709354504659995e-05,
+      "loss": 0.3385,
+      "step": 1150
+    },
+    {
+      "epoch": 1.076421248835042,
+      "grad_norm": 0.3396899881369013,
+      "learning_rate": 3.5623058336209874e-05,
+      "loss": 0.3496,
+      "step": 1155
+    },
+    {
+      "epoch": 1.0810810810810811,
+      "grad_norm": 0.28870357807743696,
+      "learning_rate": 3.553676216775975e-05,
+      "loss": 0.3406,
+      "step": 1160
+    },
+    {
+      "epoch": 1.0857409133271203,
+      "grad_norm": 0.28568156626690694,
+      "learning_rate": 3.545046599930964e-05,
+      "loss": 0.3523,
+      "step": 1165
+    },
+    {
+      "epoch": 1.0904007455731595,
+      "grad_norm": 0.27294319365310815,
+      "learning_rate": 3.536416983085951e-05,
+      "loss": 0.3537,
+      "step": 1170
+    },
+    {
+      "epoch": 1.0950605778191984,
+      "grad_norm": 0.36694769383304165,
+      "learning_rate": 3.5277873662409386e-05,
+      "loss": 0.3403,
+      "step": 1175
+    },
+    {
+      "epoch": 1.0997204100652376,
+      "grad_norm": 0.33630445896051475,
+      "learning_rate": 3.519157749395927e-05,
+      "loss": 0.3465,
+      "step": 1180
+    },
+    {
+      "epoch": 1.1043802423112767,
+      "grad_norm": 0.4121586071124287,
+      "learning_rate": 3.510528132550915e-05,
+      "loss": 0.3436,
+      "step": 1185
+    },
+    {
+      "epoch": 1.109040074557316,
+      "grad_norm": 0.3107462600176289,
+      "learning_rate": 3.501898515705903e-05,
+      "loss": 0.3455,
+      "step": 1190
+    },
+    {
+      "epoch": 1.113699906803355,
+      "grad_norm": 0.32616980094028847,
+      "learning_rate": 3.4932688988608906e-05,
+      "loss": 0.3569,
+      "step": 1195
+    },
+    {
+      "epoch": 1.1183597390493942,
+      "grad_norm": 0.5721259222478717,
+      "learning_rate": 3.4846392820158784e-05,
+      "loss": 0.3462,
+      "step": 1200
+    },
+    {
+      "epoch": 1.1230195712954334,
+      "grad_norm": 0.39091963037590294,
+      "learning_rate": 3.476009665170867e-05,
+      "loss": 0.3458,
+      "step": 1205
+    },
+    {
+      "epoch": 1.1276794035414726,
+      "grad_norm": 0.3647869778433765,
+      "learning_rate": 3.467380048325855e-05,
+      "loss": 0.3463,
+      "step": 1210
+    },
+    {
+      "epoch": 1.1323392357875117,
+      "grad_norm": 0.3278226928089527,
+      "learning_rate": 3.458750431480842e-05,
+      "loss": 0.3511,
+      "step": 1215
+    },
+    {
+      "epoch": 1.1369990680335509,
+      "grad_norm": 0.36472407535793705,
+      "learning_rate": 3.4501208146358304e-05,
+      "loss": 0.3507,
+      "step": 1220
+    },
+    {
+      "epoch": 1.14165890027959,
+      "grad_norm": 0.3162625401383028,
+      "learning_rate": 3.441491197790818e-05,
+      "loss": 0.3465,
+      "step": 1225
+    },
+    {
+      "epoch": 1.146318732525629,
+      "grad_norm": 0.2935130903926503,
+      "learning_rate": 3.432861580945806e-05,
+      "loss": 0.3474,
+      "step": 1230
+    },
+    {
+      "epoch": 1.1509785647716682,
+      "grad_norm": 0.34052282866497324,
+      "learning_rate": 3.424231964100794e-05,
+      "loss": 0.3463,
+      "step": 1235
+    },
+    {
+      "epoch": 1.1556383970177073,
+      "grad_norm": 0.36877838244359007,
+      "learning_rate": 3.415602347255782e-05,
+      "loss": 0.3453,
+      "step": 1240
+    },
+    {
+      "epoch": 1.1602982292637465,
+      "grad_norm": 0.2733712910566977,
+      "learning_rate": 3.40697273041077e-05,
+      "loss": 0.3519,
+      "step": 1245
+    },
+    {
+      "epoch": 1.1649580615097856,
+      "grad_norm": 0.37390448821752814,
+      "learning_rate": 3.398343113565758e-05,
+      "loss": 0.3457,
+      "step": 1250
+    },
+    {
+      "epoch": 1.1696178937558248,
+      "grad_norm": 0.2862398262715606,
+      "learning_rate": 3.389713496720746e-05,
+      "loss": 0.3254,
+      "step": 1255
+    },
+    {
+      "epoch": 1.174277726001864,
+      "grad_norm": 0.27627811665115726,
+      "learning_rate": 3.381083879875734e-05,
+      "loss": 0.3449,
+      "step": 1260
+    },
+    {
+      "epoch": 1.1789375582479031,
+      "grad_norm": 0.2984551216682011,
+      "learning_rate": 3.3724542630307215e-05,
+      "loss": 0.3412,
+      "step": 1265
+    },
+    {
+      "epoch": 1.1835973904939423,
+      "grad_norm": 0.3117052887160424,
+      "learning_rate": 3.363824646185709e-05,
+      "loss": 0.3501,
+      "step": 1270
+    },
+    {
+      "epoch": 1.1882572227399812,
+      "grad_norm": 0.26523368148938603,
+      "learning_rate": 3.355195029340698e-05,
+      "loss": 0.3443,
+      "step": 1275
+    },
+    {
+      "epoch": 1.1929170549860204,
+      "grad_norm": 0.3235527702018126,
+      "learning_rate": 3.346565412495685e-05,
+      "loss": 0.3544,
+      "step": 1280
+    },
+    {
+      "epoch": 1.1975768872320596,
+      "grad_norm": 0.33231065381424213,
+      "learning_rate": 3.3379357956506735e-05,
+      "loss": 0.3557,
+      "step": 1285
+    },
+    {
+      "epoch": 1.2022367194780987,
+      "grad_norm": 0.33109896147037454,
+      "learning_rate": 3.329306178805661e-05,
+      "loss": 0.3391,
+      "step": 1290
+    },
+    {
+      "epoch": 1.206896551724138,
+      "grad_norm": 0.29911644384621444,
+      "learning_rate": 3.320676561960649e-05,
+      "loss": 0.3439,
+      "step": 1295
+    },
+    {
+      "epoch": 1.211556383970177,
+      "grad_norm": 0.4166015301847235,
+      "learning_rate": 3.312046945115637e-05,
+      "loss": 0.352,
+      "step": 1300
+    },
+    {
+      "epoch": 1.2162162162162162,
+      "grad_norm": 0.40766415919978594,
+      "learning_rate": 3.303417328270625e-05,
+      "loss": 0.3509,
+      "step": 1305
+    },
+    {
+      "epoch": 1.2208760484622554,
+      "grad_norm": 0.32382906649763094,
+      "learning_rate": 3.2947877114256126e-05,
+      "loss": 0.34,
+      "step": 1310
+    },
+    {
+      "epoch": 1.2255358807082946,
+      "grad_norm": 0.30767189965326197,
+      "learning_rate": 3.286158094580601e-05,
+      "loss": 0.3602,
+      "step": 1315
+    },
+    {
+      "epoch": 1.2301957129543337,
+      "grad_norm": 0.31880798888411954,
+      "learning_rate": 3.277528477735589e-05,
+      "loss": 0.3458,
+      "step": 1320
+    },
+    {
+      "epoch": 1.2348555452003729,
+      "grad_norm": 0.31313877181562205,
+      "learning_rate": 3.268898860890577e-05,
+      "loss": 0.3526,
+      "step": 1325
+    },
+    {
+      "epoch": 1.2395153774464118,
+      "grad_norm": 0.2890512077214384,
+      "learning_rate": 3.2602692440455645e-05,
+      "loss": 0.3408,
+      "step": 1330
+    },
+    {
+      "epoch": 1.244175209692451,
+      "grad_norm": 0.2836812272018177,
+      "learning_rate": 3.2516396272005524e-05,
+      "loss": 0.3434,
+      "step": 1335
+    },
+    {
+      "epoch": 1.2488350419384902,
+      "grad_norm": 0.4540592435912553,
+      "learning_rate": 3.243010010355541e-05,
+      "loss": 0.3492,
+      "step": 1340
+    },
+    {
+      "epoch": 1.2534948741845293,
+      "grad_norm": 0.3702760451037345,
+      "learning_rate": 3.234380393510528e-05,
+      "loss": 0.3368,
+      "step": 1345
+    },
+    {
+      "epoch": 1.2581547064305685,
+      "grad_norm": 0.2917660806063631,
+      "learning_rate": 3.225750776665516e-05,
+      "loss": 0.3437,
+      "step": 1350
+    },
+    {
+      "epoch": 1.2628145386766076,
+      "grad_norm": 0.3005932483580072,
+      "learning_rate": 3.217121159820504e-05,
+      "loss": 0.3533,
+      "step": 1355
+    },
+    {
+      "epoch": 1.2674743709226468,
+      "grad_norm": 0.31926938268885113,
+      "learning_rate": 3.208491542975492e-05,
+      "loss": 0.3433,
+      "step": 1360
+    },
+    {
+      "epoch": 1.272134203168686,
+      "grad_norm": 0.31462085002303564,
+      "learning_rate": 3.19986192613048e-05,
+      "loss": 0.3484,
+      "step": 1365
+    },
+    {
+      "epoch": 1.2767940354147251,
+      "grad_norm": 0.3668129958844129,
+      "learning_rate": 3.191232309285468e-05,
+      "loss": 0.3473,
+      "step": 1370
+    },
+    {
+      "epoch": 1.281453867660764,
+      "grad_norm": 0.36011114846832903,
+      "learning_rate": 3.1826026924404556e-05,
+      "loss": 0.3374,
+      "step": 1375
+    },
+    {
+      "epoch": 1.2861136999068035,
+      "grad_norm": 0.3707136617543373,
+      "learning_rate": 3.173973075595444e-05,
+      "loss": 0.3519,
+      "step": 1380
+    },
+    {
+      "epoch": 1.2907735321528424,
+      "grad_norm": 0.31634559612246044,
+      "learning_rate": 3.165343458750431e-05,
+      "loss": 0.3532,
+      "step": 1385
+    },
+    {
+      "epoch": 1.2954333643988816,
+      "grad_norm": 0.36512675923388166,
+      "learning_rate": 3.156713841905419e-05,
+      "loss": 0.344,
+      "step": 1390
+    },
+    {
+      "epoch": 1.3000931966449207,
+      "grad_norm": 0.2724467755048477,
+      "learning_rate": 3.1480842250604076e-05,
+      "loss": 0.3483,
+      "step": 1395
+    },
+    {
+      "epoch": 1.30475302889096,
+      "grad_norm": 0.360595305928704,
+      "learning_rate": 3.1394546082153954e-05,
+      "loss": 0.3423,
+      "step": 1400
+    },
+    {
+      "epoch": 1.309412861136999,
+      "grad_norm": 0.331462059138007,
+      "learning_rate": 3.130824991370383e-05,
+      "loss": 0.3533,
+      "step": 1405
+    },
+    {
+      "epoch": 1.3140726933830382,
+      "grad_norm": 0.3123108183369381,
+      "learning_rate": 3.122195374525371e-05,
+      "loss": 0.3412,
+      "step": 1410
+    },
+    {
+      "epoch": 1.3187325256290774,
+      "grad_norm": 0.30675615259167105,
+      "learning_rate": 3.113565757680359e-05,
+      "loss": 0.3486,
+      "step": 1415
+    },
+    {
+      "epoch": 1.3233923578751166,
+      "grad_norm": 0.3247753979750987,
+      "learning_rate": 3.1049361408353474e-05,
+      "loss": 0.336,
+      "step": 1420
+    },
+    {
+      "epoch": 1.3280521901211557,
+      "grad_norm": 0.24573704166768065,
+      "learning_rate": 3.096306523990335e-05,
+      "loss": 0.3439,
+      "step": 1425
+    },
+    {
+      "epoch": 1.3327120223671947,
+      "grad_norm": 0.2784691576334795,
+      "learning_rate": 3.0876769071453223e-05,
+      "loss": 0.3523,
+      "step": 1430
+    },
+    {
+      "epoch": 1.337371854613234,
+      "grad_norm": 0.3279920083531312,
+      "learning_rate": 3.079047290300311e-05,
+      "loss": 0.3504,
+      "step": 1435
+    },
+    {
+      "epoch": 1.342031686859273,
+      "grad_norm": 0.630072707291437,
+      "learning_rate": 3.070417673455299e-05,
+      "loss": 0.3508,
+      "step": 1440
+    },
+    {
+      "epoch": 1.3466915191053122,
+      "grad_norm": 0.3062935559964653,
+      "learning_rate": 3.061788056610287e-05,
+      "loss": 0.3534,
+      "step": 1445
+    },
+    {
+      "epoch": 1.3513513513513513,
+      "grad_norm": 0.34136097201146864,
+      "learning_rate": 3.053158439765274e-05,
+      "loss": 0.3395,
+      "step": 1450
+    },
+    {
+      "epoch": 1.3560111835973905,
+      "grad_norm": 0.3051224256041293,
+      "learning_rate": 3.0445288229202625e-05,
+      "loss": 0.3469,
+      "step": 1455
+    },
+    {
+      "epoch": 1.3606710158434296,
+      "grad_norm": 0.2975038729445838,
+      "learning_rate": 3.0358992060752506e-05,
+      "loss": 0.3529,
+      "step": 1460
+    },
+    {
+      "epoch": 1.3653308480894688,
+      "grad_norm": 0.31823310979778036,
+      "learning_rate": 3.027269589230238e-05,
+      "loss": 0.347,
+      "step": 1465
+    },
+    {
+      "epoch": 1.369990680335508,
+      "grad_norm": 0.2896430196330622,
+      "learning_rate": 3.018639972385226e-05,
+      "loss": 0.3439,
+      "step": 1470
+    },
+    {
+      "epoch": 1.3746505125815471,
+      "grad_norm": 0.352875453228656,
+      "learning_rate": 3.010010355540214e-05,
+      "loss": 0.346,
+      "step": 1475
+    },
+    {
+      "epoch": 1.3793103448275863,
+      "grad_norm": 0.25417462359338994,
+      "learning_rate": 3.001380738695202e-05,
+      "loss": 0.3302,
+      "step": 1480
+    },
+    {
+      "epoch": 1.3839701770736252,
+      "grad_norm": 0.2981082945058558,
+      "learning_rate": 2.99275112185019e-05,
+      "loss": 0.3493,
+      "step": 1485
+    },
+    {
+      "epoch": 1.3886300093196646,
+      "grad_norm": 0.3626842220779335,
+      "learning_rate": 2.984121505005178e-05,
+      "loss": 0.3522,
+      "step": 1490
+    },
+    {
+      "epoch": 1.3932898415657036,
+      "grad_norm": 0.34072090786926984,
+      "learning_rate": 2.9754918881601657e-05,
+      "loss": 0.3605,
+      "step": 1495
+    },
+    {
+      "epoch": 1.3979496738117427,
+      "grad_norm": 0.36566049127137545,
+      "learning_rate": 2.966862271315154e-05,
+      "loss": 0.3572,
+      "step": 1500
+    },
+    {
+      "epoch": 1.402609506057782,
+      "grad_norm": 0.30367298168910534,
+      "learning_rate": 2.9582326544701417e-05,
+      "loss": 0.3503,
+      "step": 1505
+    },
+    {
+      "epoch": 1.407269338303821,
+      "grad_norm": 0.2828621496023037,
+      "learning_rate": 2.9496030376251292e-05,
+      "loss": 0.3454,
+      "step": 1510
+    },
+    {
+      "epoch": 1.4119291705498602,
+      "grad_norm": 0.3069391721232517,
+      "learning_rate": 2.9409734207801177e-05,
+      "loss": 0.3503,
+      "step": 1515
+    },
+    {
+      "epoch": 1.4165890027958994,
+      "grad_norm": 0.37676918497607415,
+      "learning_rate": 2.9323438039351052e-05,
+      "loss": 0.3483,
+      "step": 1520
+    },
+    {
+      "epoch": 1.4212488350419386,
+      "grad_norm": 0.3439297031393127,
+      "learning_rate": 2.9237141870900937e-05,
+      "loss": 0.3447,
+      "step": 1525
+    },
+    {
+      "epoch": 1.4259086672879777,
+      "grad_norm": 0.3359348980856527,
+      "learning_rate": 2.9150845702450812e-05,
+      "loss": 0.3434,
+      "step": 1530
+    },
+    {
+      "epoch": 1.4305684995340169,
+      "grad_norm": 0.3706215198106378,
+      "learning_rate": 2.906454953400069e-05,
+      "loss": 0.3481,
+      "step": 1535
+    },
+    {
+      "epoch": 1.4352283317800558,
+      "grad_norm": 0.26527833144434804,
+      "learning_rate": 2.897825336555057e-05,
+      "loss": 0.3443,
+      "step": 1540
+    },
+    {
+      "epoch": 1.439888164026095,
+      "grad_norm": 0.35150178535932075,
+      "learning_rate": 2.889195719710045e-05,
+      "loss": 0.3446,
+      "step": 1545
+    },
+    {
+      "epoch": 1.4445479962721341,
+      "grad_norm": 0.34068866122527197,
+      "learning_rate": 2.8805661028650328e-05,
+      "loss": 0.341,
+      "step": 1550
+    },
+    {
+      "epoch": 1.4492078285181733,
+      "grad_norm": 0.44905663737611096,
+      "learning_rate": 2.871936486020021e-05,
+      "loss": 0.3386,
+      "step": 1555
+    },
+    {
+      "epoch": 1.4538676607642125,
+      "grad_norm": 0.31579368636055327,
+      "learning_rate": 2.8633068691750088e-05,
+      "loss": 0.341,
+      "step": 1560
+    },
+    {
+      "epoch": 1.4585274930102516,
+      "grad_norm": 0.31166330874399784,
+      "learning_rate": 2.854677252329997e-05,
+      "loss": 0.3434,
+      "step": 1565
+    },
+    {
+      "epoch": 1.4631873252562908,
+      "grad_norm": 0.3677885578765502,
+      "learning_rate": 2.8460476354849848e-05,
+      "loss": 0.3502,
+      "step": 1570
+    },
+    {
+      "epoch": 1.46784715750233,
+      "grad_norm": 0.3763718138729198,
+      "learning_rate": 2.8374180186399723e-05,
+      "loss": 0.3501,
+      "step": 1575
+    },
+    {
+      "epoch": 1.4725069897483691,
+      "grad_norm": 0.3319071868875067,
+      "learning_rate": 2.8287884017949608e-05,
+      "loss": 0.3513,
+      "step": 1580
+    },
+    {
+      "epoch": 1.477166821994408,
+      "grad_norm": 0.35758179709780585,
+      "learning_rate": 2.8201587849499482e-05,
+      "loss": 0.3444,
+      "step": 1585
+    },
+    {
+      "epoch": 1.4818266542404475,
+      "grad_norm": 0.33577991103004173,
+      "learning_rate": 2.811529168104936e-05,
+      "loss": 0.3545,
+      "step": 1590
+    },
+    {
+      "epoch": 1.4864864864864864,
+      "grad_norm": 0.3025741672240685,
+      "learning_rate": 2.8028995512599242e-05,
+      "loss": 0.3453,
+      "step": 1595
+    },
+    {
+      "epoch": 1.4911463187325256,
+      "grad_norm": 0.3828036069686848,
+      "learning_rate": 2.794269934414912e-05,
+      "loss": 0.3472,
+      "step": 1600
+    },
+    {
+      "epoch": 1.4958061509785647,
+      "grad_norm": 0.3043778519192715,
+      "learning_rate": 2.7856403175699002e-05,
+      "loss": 0.3507,
+      "step": 1605
+    },
+    {
+      "epoch": 1.500465983224604,
+      "grad_norm": 0.3382233040513506,
+      "learning_rate": 2.777010700724888e-05,
+      "loss": 0.3461,
+      "step": 1610
+    },
+    {
+      "epoch": 1.505125815470643,
+      "grad_norm": 0.33027493167325966,
+      "learning_rate": 2.768381083879876e-05,
+      "loss": 0.3412,
+      "step": 1615
+    },
+    {
+      "epoch": 1.5097856477166822,
+      "grad_norm": 0.28284106409517795,
+      "learning_rate": 2.759751467034864e-05,
+      "loss": 0.3434,
+      "step": 1620
+    },
+    {
+      "epoch": 1.5144454799627214,
+      "grad_norm": 0.27285762390023743,
+      "learning_rate": 2.751121850189852e-05,
+      "loss": 0.3381,
+      "step": 1625
+    },
+    {
+      "epoch": 1.5191053122087603,
+      "grad_norm": 0.2856683184270901,
+      "learning_rate": 2.7424922333448393e-05,
+      "loss": 0.3312,
+      "step": 1630
+    },
+    {
+      "epoch": 1.5237651444547997,
+      "grad_norm": 0.32602889682943975,
+      "learning_rate": 2.7338626164998278e-05,
+      "loss": 0.3384,
+      "step": 1635
+    },
+    {
+      "epoch": 1.5284249767008387,
+      "grad_norm": 0.3112657240375754,
+      "learning_rate": 2.7252329996548153e-05,
+      "loss": 0.3412,
+      "step": 1640
+    },
+    {
+      "epoch": 1.533084808946878,
+      "grad_norm": 0.26091256859148315,
+      "learning_rate": 2.7166033828098038e-05,
+      "loss": 0.3566,
+      "step": 1645
+    },
+    {
+      "epoch": 1.537744641192917,
+      "grad_norm": 0.3123052376838657,
+      "learning_rate": 2.7079737659647913e-05,
+      "loss": 0.3455,
+      "step": 1650
+    }
+  ],
+  "logging_steps": 5,
+  "max_steps": 3219,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 3,
+  "save_steps": 550,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 1.412694515424166e+18,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-1650/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d3046f52588a443b20a3609c281264c201765ee86ad14b8ee06a6c419a206096
+size 7480

checkpoint-1650/zero_to_fp32.py

@@ -0,0 +1,674 @@
+#!/usr/bin/env python
+
+# Copyright (c) Microsoft Corporation.
+# SPDX-License-Identifier: Apache-2.0
+
+# DeepSpeed Team
+
+# This script extracts fp32 consolidated weights from a zero 1, 2 and 3 DeepSpeed checkpoints. It gets
+# copied into the top level checkpoint dir, so the user can easily do the conversion at any point in
+# the future. Once extracted, the weights don't require DeepSpeed and can be used in any
+# application.
+#
+# example:
+#   python zero_to_fp32.py . output_dir/
+#   or
+#   python zero_to_fp32.py . output_dir/ --safe_serialization
+
+import argparse
+import torch
+import glob
+import math
+import os
+import re
+import json
+from tqdm import tqdm
+from collections import OrderedDict
+from dataclasses import dataclass
+
+# while this script doesn't use deepspeed to recover data, since the checkpoints are pickled with
+# DeepSpeed data structures it has to be available in the current python environment.
+from deepspeed.utils import logger
+from deepspeed.checkpoint.constants import (DS_VERSION, OPTIMIZER_STATE_DICT, SINGLE_PARTITION_OF_FP32_GROUPS,
+                                            FP32_FLAT_GROUPS, ZERO_STAGE, PARTITION_COUNT, PARAM_SHAPES, BUFFER_NAMES,
+                                            FROZEN_PARAM_SHAPES, FROZEN_PARAM_FRAGMENTS)
+
+
+@dataclass
+class zero_model_state:
+    buffers: dict()
+    param_shapes: dict()
+    shared_params: list
+    ds_version: int
+    frozen_param_shapes: dict()
+    frozen_param_fragments: dict()
+
+
+debug = 0
+
+# load to cpu
+device = torch.device('cpu')
+
+
+def atoi(text):
+    return int(text) if text.isdigit() else text
+
+
+def natural_keys(text):
+    '''
+    alist.sort(key=natural_keys) sorts in human order
+    http://nedbatchelder.com/blog/200712/human_sorting.html
+    (See Toothy's implementation in the comments)
+    '''
+    return [atoi(c) for c in re.split(r'(\d+)', text)]
+
+
+def get_model_state_file(checkpoint_dir, zero_stage):
+    if not os.path.isdir(checkpoint_dir):
+        raise FileNotFoundError(f"Directory '{checkpoint_dir}' doesn't exist")
+
+    # there should be only one file
+    if zero_stage <= 2:
+        file = os.path.join(checkpoint_dir, "mp_rank_00_model_states.pt")
+    elif zero_stage == 3:
+        file = os.path.join(checkpoint_dir, "zero_pp_rank_0_mp_rank_00_model_states.pt")
+
+    if not os.path.exists(file):
+        raise FileNotFoundError(f"can't find model states file at '{file}'")
+
+    return file
+
+
+def get_checkpoint_files(checkpoint_dir, glob_pattern):
+    # XXX: need to test that this simple glob rule works for multi-node setup too
+    ckpt_files = sorted(glob.glob(os.path.join(checkpoint_dir, glob_pattern)), key=natural_keys)
+
+    if len(ckpt_files) == 0:
+        raise FileNotFoundError(f"can't find {glob_pattern} files in directory '{checkpoint_dir}'")
+
+    return ckpt_files
+
+
+def get_optim_files(checkpoint_dir):
+    return get_checkpoint_files(checkpoint_dir, "*_optim_states.pt")
+
+
+def get_model_state_files(checkpoint_dir):
+    return get_checkpoint_files(checkpoint_dir, "*_model_states.pt")
+
+
+def parse_model_states(files):
+    zero_model_states = []
+    for file in files:
+        state_dict = torch.load(file, map_location=device)
+
+        if BUFFER_NAMES not in state_dict:
+            raise ValueError(f"{file} is not a model state checkpoint")
+        buffer_names = state_dict[BUFFER_NAMES]
+        if debug:
+            print("Found buffers:", buffer_names)
+
+        # recover just the buffers while restoring them to fp32 if they were saved in fp16
+        buffers = {k: v.float() for k, v in state_dict["module"].items() if k in buffer_names}
+        param_shapes = state_dict[PARAM_SHAPES]
+
+        # collect parameters that are included in param_shapes
+        param_names = []
+        for s in param_shapes:
+            for name in s.keys():
+                param_names.append(name)
+
+        # update with frozen parameters
+        frozen_param_shapes = state_dict.get(FROZEN_PARAM_SHAPES, None)
+        if frozen_param_shapes is not None:
+            if debug:
+                print(f"Found frozen_param_shapes: {frozen_param_shapes}")
+            param_names += list(frozen_param_shapes.keys())
+
+        # handle shared params
+        shared_params = [[k, v] for k, v in state_dict["shared_params"].items()]
+
+        ds_version = state_dict.get(DS_VERSION, None)
+
+        frozen_param_fragments = state_dict.get(FROZEN_PARAM_FRAGMENTS, None)
+
+        z_model_state = zero_model_state(buffers=buffers,
+                                         param_shapes=param_shapes,
+                                         shared_params=shared_params,
+                                         ds_version=ds_version,
+                                         frozen_param_shapes=frozen_param_shapes,
+                                         frozen_param_fragments=frozen_param_fragments)
+        zero_model_states.append(z_model_state)
+
+    return zero_model_states
+
+
+def parse_optim_states(files, ds_checkpoint_dir):
+    total_files = len(files)
+    state_dicts = []
+    for f in files:
+        state_dict = torch.load(f, map_location=device)
+        # immediately discard the potentially huge 2 optimizer states as we only care for fp32 master weights
+        # and also handle the case where it was already removed by another helper script
+        state_dict["optimizer_state_dict"].pop("optimizer_state_dict", None)
+        state_dicts.append(state_dict)
+
+    if not ZERO_STAGE in state_dicts[0][OPTIMIZER_STATE_DICT]:
+        raise ValueError(f"{files[0]} is not a zero checkpoint")
+    zero_stage = state_dicts[0][OPTIMIZER_STATE_DICT][ZERO_STAGE]
+    world_size = state_dicts[0][OPTIMIZER_STATE_DICT][PARTITION_COUNT]
+
+    # For ZeRO-2 each param group can have different partition_count as data parallelism for expert
+    # parameters can be different from data parallelism for non-expert parameters. So we can just
+    # use the max of the partition_count to get the dp world_size.
+
+    if type(world_size) is list:
+        world_size = max(world_size)
+
+    if world_size != total_files:
+        raise ValueError(
+            f"Expected {world_size} of '*_optim_states.pt' under '{ds_checkpoint_dir}' but found {total_files} files. "
+            "Possibly due to an overwrite of an old checkpoint, or a checkpoint didn't get saved by one or more processes."
+        )
+
+    # the groups are named differently in each stage
+    if zero_stage <= 2:
+        fp32_groups_key = SINGLE_PARTITION_OF_FP32_GROUPS
+    elif zero_stage == 3:
+        fp32_groups_key = FP32_FLAT_GROUPS
+    else:
+        raise ValueError(f"unknown zero stage {zero_stage}")
+
+    if zero_stage <= 2:
+        fp32_flat_groups = [state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key] for i in range(len(state_dicts))]
+    elif zero_stage == 3:
+        # if there is more than one param group, there will be multiple flattened tensors - one
+        # flattened tensor per group - for simplicity merge them into a single tensor
+        #
+        # XXX: could make the script more memory efficient for when there are multiple groups - it
+        # will require matching the sub-lists of param_shapes for each param group flattened tensor
+
+        fp32_flat_groups = [
+            torch.cat(state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key], 0) for i in range(len(state_dicts))
+        ]
+
+    return zero_stage, world_size, fp32_flat_groups
+
+
+def _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters):
+    """
+    Returns fp32 state_dict reconstructed from ds checkpoint
+
+    Args:
+        - ``ds_checkpoint_dir``: path to the deepspeed checkpoint folder (where the optimizer files are)
+
+    """
+    print(f"Processing zero checkpoint '{ds_checkpoint_dir}'")
+
+    optim_files = get_optim_files(ds_checkpoint_dir)
+    zero_stage, world_size, fp32_flat_groups = parse_optim_states(optim_files, ds_checkpoint_dir)
+    print(f"Detected checkpoint of type zero stage {zero_stage}, world_size: {world_size}")
+
+    model_files = get_model_state_files(ds_checkpoint_dir)
+
+    zero_model_states = parse_model_states(model_files)
+    print(f'Parsing checkpoint created by deepspeed=={zero_model_states[0].ds_version}')
+
+    if zero_stage <= 2:
+        return _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states,
+                                                          exclude_frozen_parameters)
+    elif zero_stage == 3:
+        return _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states,
+                                                          exclude_frozen_parameters)
+
+
+def _zero2_merge_frozen_params(state_dict, zero_model_states):
+    if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0:
+        return
+
+    frozen_param_shapes = zero_model_states[0].frozen_param_shapes
+    frozen_param_fragments = zero_model_states[0].frozen_param_fragments
+
+    if debug:
+        num_elem = sum(s.numel() for s in frozen_param_shapes.values())
+        print(f'rank 0: {FROZEN_PARAM_SHAPES}.numel = {num_elem}')
+
+        wanted_params = len(frozen_param_shapes)
+        wanted_numel = sum(s.numel() for s in frozen_param_shapes.values())
+        avail_numel = sum([p.numel() for p in frozen_param_fragments.values()])
+        print(f'Frozen params: Have {avail_numel} numels to process.')
+        print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params')
+
+    total_params = 0
+    total_numel = 0
+    for name, shape in frozen_param_shapes.items():
+        total_params += 1
+        unpartitioned_numel = shape.numel()
+        total_numel += unpartitioned_numel
+
+        state_dict[name] = frozen_param_fragments[name]
+
+        if debug:
+            print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ")
+
+    print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements")
+
+
+def _has_callable(obj, fn):
+    attr = getattr(obj, fn, None)
+    return callable(attr)
+
+
+def _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states):
+    param_shapes = zero_model_states[0].param_shapes
+
+    # Reconstruction protocol:
+    #
+    # XXX: document this
+
+    if debug:
+        for i in range(world_size):
+            for j in range(len(fp32_flat_groups[0])):
+                print(f"{FP32_FLAT_GROUPS}[{i}][{j}].shape={fp32_flat_groups[i][j].shape}")
+
+    # XXX: memory usage doubles here (zero2)
+    num_param_groups = len(fp32_flat_groups[0])
+    merged_single_partition_of_fp32_groups = []
+    for i in range(num_param_groups):
+        merged_partitions = [sd[i] for sd in fp32_flat_groups]
+        full_single_fp32_vector = torch.cat(merged_partitions, 0)
+        merged_single_partition_of_fp32_groups.append(full_single_fp32_vector)
+    avail_numel = sum(
+        [full_single_fp32_vector.numel() for full_single_fp32_vector in merged_single_partition_of_fp32_groups])
+
+    if debug:
+        wanted_params = sum([len(shapes) for shapes in param_shapes])
+        wanted_numel = sum([sum(shape.numel() for shape in shapes.values()) for shapes in param_shapes])
+        # not asserting if there is a mismatch due to possible padding
+        print(f"Have {avail_numel} numels to process.")
+        print(f"Need {wanted_numel} numels in {wanted_params} params.")
+
+    # params
+    # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support
+    # out-of-core computing solution
+    total_numel = 0
+    total_params = 0
+    for shapes, full_single_fp32_vector in zip(param_shapes, merged_single_partition_of_fp32_groups):
+        offset = 0
+        avail_numel = full_single_fp32_vector.numel()
+        for name, shape in shapes.items():
+
+            unpartitioned_numel = shape.numel() if _has_callable(shape, 'numel') else math.prod(shape)
+            total_numel += unpartitioned_numel
+            total_params += 1
+
+            if debug:
+                print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ")
+            state_dict[name] = full_single_fp32_vector.narrow(0, offset, unpartitioned_numel).view(shape)
+            offset += unpartitioned_numel
+
+        # Z2 started to align to 2*world_size to improve nccl performance. Therefore both offset and
+        # avail_numel can differ by anywhere between 0..2*world_size. Due to two unrelated complex
+        # paddings performed in the code it's almost impossible to predict the exact numbers w/o the
+        # live optimizer object, so we are checking that the numbers are within the right range
+        align_to = 2 * world_size
+
+        def zero2_align(x):
+            return align_to * math.ceil(x / align_to)
+
+        if debug:
+            print(f"original offset={offset}, avail_numel={avail_numel}")
+
+        offset = zero2_align(offset)
+        avail_numel = zero2_align(avail_numel)
+
+        if debug:
+            print(f"aligned  offset={offset}, avail_numel={avail_numel}")
+
+        # Sanity check
+        if offset != avail_numel:
+            raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong")
+
+    print(f"Reconstructed fp32 state dict with {total_params} params {total_numel} elements")
+
+
+def _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states,
+                                               exclude_frozen_parameters):
+    state_dict = OrderedDict()
+
+    # buffers
+    buffers = zero_model_states[0].buffers
+    state_dict.update(buffers)
+    if debug:
+        print(f"added {len(buffers)} buffers")
+
+    if not exclude_frozen_parameters:
+        _zero2_merge_frozen_params(state_dict, zero_model_states)
+
+    _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states)
+
+    # recover shared parameters
+    for pair in zero_model_states[0].shared_params:
+        if pair[1] in state_dict:
+            state_dict[pair[0]] = state_dict[pair[1]]
+
+    return state_dict
+
+
+def zero3_partitioned_param_info(unpartitioned_numel, world_size):
+    remainder = unpartitioned_numel % world_size
+    padding_numel = (world_size - remainder) if remainder else 0
+    partitioned_numel = math.ceil(unpartitioned_numel / world_size)
+    return partitioned_numel, padding_numel
+
+
+def _zero3_merge_frozen_params(state_dict, world_size, zero_model_states):
+    if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0:
+        return
+
+    if debug:
+        for i in range(world_size):
+            num_elem = sum(s.numel() for s in zero_model_states[i].frozen_param_fragments.values())
+            print(f'rank {i}: {FROZEN_PARAM_SHAPES}.numel = {num_elem}')
+
+        frozen_param_shapes = zero_model_states[0].frozen_param_shapes
+        wanted_params = len(frozen_param_shapes)
+        wanted_numel = sum(s.numel() for s in frozen_param_shapes.values())
+        avail_numel = sum([p.numel() for p in zero_model_states[0].frozen_param_fragments.values()]) * world_size
+        print(f'Frozen params: Have {avail_numel} numels to process.')
+        print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params')
+
+    total_params = 0
+    total_numel = 0
+    for name, shape in zero_model_states[0].frozen_param_shapes.items():
+        total_params += 1
+        unpartitioned_numel = shape.numel()
+        total_numel += unpartitioned_numel
+
+        param_frags = tuple(model_state.frozen_param_fragments[name] for model_state in zero_model_states)
+        state_dict[name] = torch.cat(param_frags, 0).narrow(0, 0, unpartitioned_numel).view(shape)
+
+        partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size)
+
+        if debug:
+            print(
+                f"Frozen params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}"
+            )
+
+    print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements")
+
+
+def _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states):
+    param_shapes = zero_model_states[0].param_shapes
+    avail_numel = fp32_flat_groups[0].numel() * world_size
+    # Reconstruction protocol: For zero3 we need to zip the partitions together at boundary of each
+    # param, re-consolidating each param, while dealing with padding if any
+
+    # merge list of dicts, preserving order
+    param_shapes = {k: v for d in param_shapes for k, v in d.items()}
+
+    if debug:
+        for i in range(world_size):
+            print(f"{FP32_FLAT_GROUPS}[{i}].shape={fp32_flat_groups[i].shape}")
+
+        wanted_params = len(param_shapes)
+        wanted_numel = sum(shape.numel() for shape in param_shapes.values())
+        # not asserting if there is a mismatch due to possible padding
+        avail_numel = fp32_flat_groups[0].numel() * world_size
+        print(f"Trainable params: Have {avail_numel} numels to process.")
+        print(f"Trainable params: Need {wanted_numel} numels in {wanted_params} params.")
+
+    # params
+    # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support
+    # out-of-core computing solution
+    offset = 0
+    total_numel = 0
+    total_params = 0
+    for name, shape in tqdm(param_shapes.items(), desc='Gathering Sharded Weights'):
+        unpartitioned_numel = shape.numel()
+        total_numel += unpartitioned_numel
+        total_params += 1
+        partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size)
+
+        if debug:
+            print(
+                f"Trainable params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}"
+            )
+
+        # XXX: memory usage doubles here
+        state_dict[name] = torch.cat(
+            tuple(fp32_flat_groups[i].narrow(0, offset, partitioned_numel) for i in range(world_size)),
+            0).narrow(0, 0, unpartitioned_numel).view(shape)
+        offset += partitioned_numel
+
+    offset *= world_size
+
+    # Sanity check
+    if offset != avail_numel:
+        raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong")
+
+    print(f"Reconstructed Trainable fp32 state dict with {total_params} params {total_numel} elements")
+
+
+def _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states,
+                                               exclude_frozen_parameters):
+    state_dict = OrderedDict()
+
+    # buffers
+    buffers = zero_model_states[0].buffers
+    state_dict.update(buffers)
+    if debug:
+        print(f"added {len(buffers)} buffers")
+
+    if not exclude_frozen_parameters:
+        _zero3_merge_frozen_params(state_dict, world_size, zero_model_states)
+
+    _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states)
+
+    # recover shared parameters
+    for pair in zero_model_states[0].shared_params:
+        if pair[1] in state_dict:
+            state_dict[pair[0]] = state_dict[pair[1]]
+
+    return state_dict
+
+
+def get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag=None, exclude_frozen_parameters=False):
+    """
+    Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated state_dict that can be loaded with
+    ``load_state_dict()`` and used for training without DeepSpeed or shared with others, for example
+    via a model hub.
+
+    Args:
+        - ``checkpoint_dir``: path to the desired checkpoint folder
+        - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in 'latest' file. e.g., ``global_step14``
+        - ``exclude_frozen_parameters``: exclude frozen parameters
+
+    Returns:
+        - pytorch ``state_dict``
+
+    Note: this approach may not work if your application doesn't have sufficient free CPU memory and
+    you may need to use the offline approach using the ``zero_to_fp32.py`` script that is saved with
+    the checkpoint.
+
+    A typical usage might be ::
+
+        from deepspeed.utils.zero_to_fp32 import get_fp32_state_dict_from_zero_checkpoint
+        # do the training and checkpoint saving
+        state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir) # already on cpu
+        model = model.cpu() # move to cpu
+        model.load_state_dict(state_dict)
+        # submit to model hub or save the model to share with others
+
+    In this example the ``model`` will no longer be usable in the deepspeed context of the same
+    application. i.e. you will need to re-initialize the deepspeed engine, since
+    ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it.
+
+    If you want it all done for you, use ``load_state_dict_from_zero_checkpoint`` instead.
+
+    """
+    if tag is None:
+        latest_path = os.path.join(checkpoint_dir, 'latest')
+        if os.path.isfile(latest_path):
+            with open(latest_path, 'r') as fd:
+                tag = fd.read().strip()
+        else:
+            raise ValueError(f"Unable to find 'latest' file at {latest_path}")
+
+    ds_checkpoint_dir = os.path.join(checkpoint_dir, tag)
+
+    if not os.path.isdir(ds_checkpoint_dir):
+        raise FileNotFoundError(f"Directory '{ds_checkpoint_dir}' doesn't exist")
+
+    return _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters)
+
+
+def convert_zero_checkpoint_to_fp32_state_dict(checkpoint_dir,
+                                               output_dir,
+                                               max_shard_size="5GB",
+                                               safe_serialization=False,
+                                               tag=None,
+                                               exclude_frozen_parameters=False):
+    """
+    Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict`` file that can be
+    loaded with ``torch.load(file)`` + ``load_state_dict()`` and used for training without DeepSpeed.
+
+    Args:
+        - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``)
+        - ``output_dir``: directory to the pytorch fp32 state_dict output files
+        - ``max_shard_size``: the maximum size for a checkpoint before being sharded, default value is 5GB
+        - ``safe_serialization``:  whether to save the model using `safetensors` or the traditional PyTorch way (that uses `pickle`).
+        - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14``
+        - ``exclude_frozen_parameters``: exclude frozen parameters
+    """
+    # Dependency pre-check
+    if safe_serialization:
+        try:
+            from safetensors.torch import save_file
+        except ImportError:
+            print('If you want to use `safe_serialization`, please `pip install safetensors`')
+            raise
+    if max_shard_size is not None:
+        try:
+            from huggingface_hub import split_torch_state_dict_into_shards
+        except ImportError:
+            print('If you want to use `max_shard_size`, please `pip install huggingface_hub`')
+            raise
+
+    # Convert zero checkpoint to state_dict
+    state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag, exclude_frozen_parameters)
+
+    # Shard the model if it is too big.
+    weights_name = "model.safetensors" if safe_serialization else "pytorch_model.bin"
+    if max_shard_size is not None:
+        filename_pattern = weights_name.replace(".bin", "{suffix}.bin").replace(".safetensors", "{suffix}.safetensors")
+        state_dict_split = split_torch_state_dict_into_shards(state_dict,
+                                                              filename_pattern=filename_pattern,
+                                                              max_shard_size=max_shard_size)
+    else:
+        from collections import namedtuple
+        StateDictSplit = namedtuple("StateDictSplit", ["is_sharded", "filename_to_tensors"])
+        state_dict_split = StateDictSplit(is_sharded=False,
+                                          filename_to_tensors={weights_name: list(state_dict.keys())})
+
+    # Save the model
+    filename_to_tensors = state_dict_split.filename_to_tensors.items()
+    for shard_file, tensors in tqdm(filename_to_tensors, desc="Saving checkpoint shards"):
+        shard = {tensor: state_dict[tensor].contiguous() for tensor in tensors}
+        output_path = os.path.join(output_dir, shard_file)
+        if safe_serialization:
+            save_file(shard, output_path, metadata={"format": "pt"})
+        else:
+            torch.save(shard, output_path)
+
+    # Save index if sharded
+    if state_dict_split.is_sharded:
+        index = {
+            "metadata": state_dict_split.metadata,
+            "weight_map": state_dict_split.tensor_to_filename,
+        }
+        save_index_file = "model.safetensors.index.json" if safe_serialization else "pytorch_model.bin.index.json"
+        save_index_file = os.path.join(output_dir, save_index_file)
+        with open(save_index_file, "w", encoding="utf-8") as f:
+            content = json.dumps(index, indent=2, sort_keys=True) + "\n"
+            f.write(content)
+
+
+def load_state_dict_from_zero_checkpoint(model, checkpoint_dir, tag=None):
+    """
+    1. Put the provided model to cpu
+    2. Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict``
+    3. Load it into the provided model
+
+    Args:
+        - ``model``: the model object to update
+        - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``)
+        - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14``
+
+    Returns:
+        - ``model`: modified model
+
+    Make sure you have plenty of CPU memory available before you call this function. If you don't
+    have enough use the ``zero_to_fp32.py`` utility to do the conversion. You will find it
+    conveniently placed for you in the checkpoint folder.
+
+    A typical usage might be ::
+
+        from deepspeed.utils.zero_to_fp32 import load_state_dict_from_zero_checkpoint
+        model = load_state_dict_from_zero_checkpoint(trainer.model, checkpoint_dir)
+        # submit to model hub or save the model to share with others
+
+    Note, that once this was run, the ``model`` will no longer be usable in the deepspeed context
+    of the same application. i.e. you will need to re-initialize the deepspeed engine, since
+    ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it.
+
+    """
+    logger.info(f"Extracting fp32 weights")
+    state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag)
+
+    logger.info(f"Overwriting model with fp32 weights")
+    model = model.cpu()
+    model.load_state_dict(state_dict, strict=False)
+
+    return model
+
+
+if __name__ == "__main__":
+    parser = argparse.ArgumentParser()
+    parser.add_argument("checkpoint_dir",
+                        type=str,
+                        help="path to the desired checkpoint folder, e.g., path/checkpoint-12")
+    parser.add_argument("output_dir",
+                        type=str,
+                        help="directory to the pytorch fp32 state_dict output files"
+                        "(e.g. path/checkpoint-12-output/)")
+    parser.add_argument(
+        "--max_shard_size",
+        type=str,
+        default="5GB",
+        help="The maximum size for a checkpoint before being sharded. Checkpoints shard will then be each of size"
+        "lower than this size. If expressed as a string, needs to be digits followed by a unit (like `5MB`"
+        "We default it to 5GB in order for models to be able to run easily on free-tier google colab instances"
+        "without CPU OOM issues.")
+    parser.add_argument(
+        "--safe_serialization",
+        default=False,
+        action='store_true',
+        help="Whether to save the model using `safetensors` or the traditional PyTorch way (that uses `pickle`).")
+    parser.add_argument("-t",
+                        "--tag",
+                        type=str,
+                        default=None,
+                        help="checkpoint tag used as a unique identifier for checkpoint. e.g., global_step1")
+    parser.add_argument("--exclude_frozen_parameters", action='store_true', help="exclude frozen parameters")
+    parser.add_argument("-d", "--debug", action='store_true', help="enable debug")
+    args = parser.parse_args()
+
+    debug = args.debug
+
+    convert_zero_checkpoint_to_fp32_state_dict(args.checkpoint_dir,
+                                               args.output_dir,
+                                               max_shard_size=args.max_shard_size,
+                                               safe_serialization=args.safe_serialization,
+                                               tag=args.tag,
+                                               exclude_frozen_parameters=args.exclude_frozen_parameters)