Upload model

config.json

@@ -1,5 +1,5 @@
 {
-  "_name_or_path": "/pretraining/local-models/training/2024-01-17_05-40-28",
+  "_name_or_path": "/pretraining/local-models/training/2024-01-17_05-57-45",
   "activation_function": "gelu_new",
   "architectures": [
     "GPT2LMHeadModel"
@@ -11,11 +11,10 @@
   "initializer_range": 0.02,
   "layer_norm_epsilon": 1e-05,
   "model_type": "gpt2",
-  "n_ctx": 1024,
-  "n_embd": 1250,
+  "n_embd": 760,
   "n_head": 10,
   "n_inner": null,
-  "n_layer": 9,
+  "n_layer": 12,
   "n_positions": 1024,
   "reorder_and_upcast_attn": false,
   "resid_pdrop": 0.1,

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:fcd8d39e0f6917ed5d17764b80e1296768538b5fb9a963d800e2a012d4937037
-size 932011472
+oid sha256:da2e26af1e0c683dfe84ac380b9b730b519f616fd10c1ff7b9b59395f838e458
+size 489087136