Upload model

config.json

@@ -1,27 +1,27 @@
 {
-  "_name_or_path": "/pretraining/local-models/training/2024-01-06_07-37-46",
+  "_name_or_path": "/Pretraining-Testing/pretraining/local-models/training/2024-01-08_15-58-03",
   "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:b8ca10639e9674a022d00abc4f4e01eb1586f31f34fbb703873b10dbd79055a5
-size 489087136
+oid sha256:28b036d5cc994008ae93ed0b1214783a2fcbc29a7860d2502618a6a2d4c6f7a3
+size 469421488