Upload model

config.json

@@ -1,28 +1,27 @@
 {
-  "_name_or_path": "/pretraining/local-models/training/2024-01-17_06-02-35",
+  "_name_or_path": "/pretraining/local-models/training/2024-01-17_06-13-20",
   "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_ctx": 1024,
-  "n_embd": 1450,
-  "n_head": 10,
+  "n_embd": 768,
+  "n_head": 12,
   "n_inner": null,
-  "n_layer": 10,
+  "n_layer": 20,
   "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:c5f7bc8e01685ef6ab7d94e184c197a77d5f1950b959f4b0d334d347ad0e11cf
-size 1307408120
+oid sha256:9292524a7d24022612a14f4f5b2a891a947451b6a5cbd66b3f7c288199f96793
+size 724595976