medium model 학습습, rev3

config.json

@@ -1,4 +1,5 @@
 {
+  "_name_or_path": "heegyu/ajoublue-gpt2-medium",
   "activation_function": "gelu_new",
   "architectures": [
     "GPT2LMHeadModel"
@@ -13,9 +14,13 @@
   "n_ctx": 1024,
   "n_embd": 1024,
   "n_head": 16,
+  "n_inner": null,
   "n_layer": 24,
   "n_positions": 1024,
+  "reorder_and_upcast_attn": false,
   "resid_pdrop": 0.1,
+  "scale_attn_by_inverse_layer_idx": false,
+  "scale_attn_weights": true,
   "summary_activation": null,
   "summary_first_dropout": 0.1,
   "summary_proj_to_labels": true,
@@ -27,5 +32,7 @@
       "max_length": 50
     }
   },
+  "transformers_version": "4.25.1",
+  "use_cache": true,
   "vocab_size": 51200
-}
+}

flax_model.msgpack

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:04c7640c4ef6bd5097771047737aba334669b00c73dda0c7dc5a50ae0936618c
+size 1423164830

tokenizer_config.json

@@ -19,7 +19,7 @@
   },
   "errors": "replace",
   "model_max_length": 1000000000000000019884624838656,
-  "name_or_path": "heegyu/ajoublue-gpt2-base",
+  "name_or_path": "heegyu/ajoublue-gpt2-medium",
   "pad_token": null,
   "special_tokens_map_file": null,
   "tokenizer_class": "GPT2Tokenizer",