Upload TransformerForCausalLM

config.json

@@ -1,41 +1,32 @@
 {
-  "_name_or_path": "/network/scratch/z/zhixuan.lin/linear-rnn-torch/release/fox-pro-760m-longcrawl64-48b",
+  "_name_or_path": "/network/scratch/z/zhixuan.lin/linear-rnn-torch/release/transformer-llama-760m-longcrawl64-48b",
   "architectures": [
-    "ForgettingTransformerForCausalLM"
+    "TransformerForCausalLM"
   ],
   "attention_bias": false,
   "bos_token_id": null,
-  "decay_time_max": null,
-  "decay_time_min": null,
   "elementwise_affine": true,
   "eos_token_id": null,
-  "fgate_bias_init": false,
-  "fgate_type": "full",
   "fuse_cross_entropy": true,
   "fuse_norm": true,
   "hidden_act": "swish",
-  "hidden_ratio": 3.5,
+  "hidden_ratio": 4,
   "hidden_size": 1536,
   "initializer_range": 0.02,
   "intermediate_size": null,
   "max_position_embeddings": null,
-  "model_type": "forgetting_transformer-project_fox",
+  "model_type": "transformer-project_fox",
   "norm_eps": 1e-06,
-  "num_heads": 24,
+  "num_heads": 12,
   "num_hidden_layers": 24,
   "num_kv_heads": null,
-  "ogate_act": "sigmoid",
-  "qk_norm": true,
+  "qk_norm": false,
   "rope_base": 500000.0,
   "tie_word_embeddings": false,
   "torch_dtype": "float32",
   "transformers_version": "4.44.0",
   "use_cache": true,
-  "use_k_shift": true,
-  "use_output_gate": true,
-  "use_output_norm": true,
-  "use_rope": false,
-  "use_v_shift": true,
+  "use_rope": true,
   "vocab_size": 50257,
   "window_size": null
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1eece24e2dcee0c560f576273590ff77ed93eadccfa045270ef514abe5e58124
-size 3346590184
+oid sha256:e0c8ff874672c54f6a27cd7f3d5ca8d060e6c115b5be38b95c7a7ef8c71b5eb1
+size 3335789480