checkpoint-225000 contents

config.json

@@ -1,7 +1,8 @@
 {
+  "_name_or_path": "stanford-crfm/celebrimbor-gpt2-medium-x81",
   "activation_function": "gelu_new",
   "architectures": [
-    "MistralGPT2LMHeadModel"
+    "GPT2LMHeadModel"
   ],
   "attn_pdrop": 0.1,
   "bos_token_id": 50256,
@@ -19,7 +20,10 @@
   "n_positions": 1024,
   "n_special": 0,
   "predict_special_tokens": true,
+  "reorder_and_upcast_attn": true,
   "resid_pdrop": 0.1,
+  "scale_attn_by_inverse_layer_idx": true,
+  "scale_attn_weights": true,
   "summary_activation": null,
   "summary_first_dropout": 0.1,
   "summary_proj_to_labels": true,
@@ -31,7 +35,7 @@
       "max_length": 50
     }
   },
-  "transformers_version": "4.5.0",
+  "transformers_version": "4.12.5",
   "use_cache": false,
   "vocab_size": 50257
 }

latest

@@ -1 +1 @@
-global_step224011
+global_step225011

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:bb1387226a204beb240f551b88f183d99903f7fca6a85e1a0fb95b7fdaf5a242
+oid sha256:eedc682dfbf68e284ce780a6e74b8ee35de86847a6d60f007c8294ffa85a40a5
 size 734885928

trainer_state.json

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ab15f7b14dd53416bc7ad71fa88e82e50a944cde9da707484f962e51dba0c31d
-size 15668929
+oid sha256:c493251f1ae70afa04ce3ebce39ddc567b0cdee84e6ef527112bba84652d5361
+size 15738950