Run 4. Outer Step 23. Inner Step 18.

config.json

@@ -268,24 +268,18 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5531430,
-    5531431,
-    5531434,
-    5531435,
-    5531437,
-    5531439,
-    5531441,
-    5531444,
-    5531446,
-    5531448,
-    5531451
+    5531453,
+    5531455,
+    5531458,
+    5531460,
+    5531461
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 17,
+  "inner_step": 18,
   "inner_steps": 0,
   "last_allreduce_block": 5529444,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:19506d105cb4e2cba6c4bcf48ce456054cbbe029dee72a2b5d1cefd9dd99a177
+oid sha256:62ab6c18919ed296c1a1e5fbaa2ea4f2a06d958786a49cda5444b38ee6860d47
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:33f0d00bfc2e425162951fb23166bdc414f2abc3ec6e2d4a07405cf58374932d
+oid sha256:df8f3359fdc75e82c20c78cb5fbbfa02ccc55d454503765db88e1d541d1d24ea
 size 4040701744