Run 3. Outer Step 9. Inner Step 15.

config.json

@@ -268,18 +268,18 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5399427,
-    5399431,
-    5399436,
-    5399440,
-    5399444
+    5399474,
+    5399478,
+    5399482,
+    5399486,
+    5399490
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 13,
+  "inner_step": 15,
   "inner_steps": 0,
   "last_allreduce_block": 5398844,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:489df0876aab00935744d44537d94f028c9ce8783b4accf2a70e191effe209cb
+oid sha256:9ec3313d31b63c55b31ea8bfb3e101f9ef3de75cb9000faf2de59b5e40b4f925
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:516e87e351ad9185d564f6cac7e38ee25ee31fc4e1b8e25f751ff4e675eb547b
+oid sha256:08d597234be44f9323ab5a7b113d56670d2f3d7c77becd951cda08700ad59d20
 size 4040701744