Run 4. Outer Step 40. Inner Step 38.

config.json

@@ -268,18 +268,19 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5556279,
-    5556283,
-    5556288,
-    5556293,
-    5556297
+    5556302,
+    5556306,
+    5556310,
+    5556314,
+    5556319,
+    5556323
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 37,
+  "inner_step": 38,
   "inner_steps": 0,
   "last_allreduce_block": 5555115,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7fe6237ec87f9cad428bbcae566c47cffd44d958a0824b035d1fe4338cb4ffa9
+oid sha256:14e032df4985f312aba5fa75c73175e6ce7e31c8aac47c49e9097e60c61124d8
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b6a4492e7d4e31c2ea9b33c5369cd572b8b5f4a5671a888519a0921a03728d41
+oid sha256:8f7f4c2a27dab28b570741fe0dc593f228ac18b820641dfb391dbe67f1f4948c
 size 4040701744