Upload model

config.json

@@ -1,27 +1,27 @@
 {
-  "_name_or_path": "/workspace/new/pretraining/local-models/training/2024-01-05_07-36-44",
+  "_name_or_path": "/workspace/new/pretraining/local-models/training/2024-01-06_15-31-54",
   "activation_function": "gelu_new",
   "architectures": [
     "GPT2LMHeadModel"
   ],
-  "attn_pdrop": 0.1,
+  "attn_pdrop": 0.0,
   "bos_token_id": 50256,
-  "embd_pdrop": 0.1,
+  "embd_pdrop": 0.0,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
   "layer_norm_epsilon": 1e-05,
   "model_type": "gpt2",
-  "n_embd": 760,
-  "n_head": 10,
+  "n_embd": 768,
+  "n_head": 12,
   "n_inner": null,
-  "n_layer": 12,
+  "n_layer": 11,
   "n_positions": 1024,
   "reorder_and_upcast_attn": false,
-  "resid_pdrop": 0.1,
+  "resid_pdrop": 0.0,
   "scale_attn_by_inverse_layer_idx": false,
   "scale_attn_weights": true,
   "summary_activation": null,
-  "summary_first_dropout": 0.1,
+  "summary_first_dropout": 0.0,
   "summary_proj_to_labels": true,
   "summary_type": "cls_index",
   "summary_use_proj": true,

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9db148a9312dfd716353b103e54b11c0ba7a700b4efd7d04e8bb0f2a9e7c3cf3
-size 489087136
+oid sha256:4c44d7deda32f22e5773f7974eddc994e2c5aba8cea7e1f3e902efbd0836b572
+size 469421488