Run 3. Outer Step 6. Inner Step 258.

config.json

@@ -268,18 +268,18 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5389261,
-    5389265,
-    5389268,
-    5389272,
-    5389276
+    5389308,
+    5389311,
+    5389315,
+    5389319,
+    5389322
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 255,
+  "inner_step": 258,
   "inner_steps": 0,
   "last_allreduce_block": 5378990,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8e5c88552a77135dd43d2a9a60087061daa748b5ee1b7a875646b48c4a3881db
+oid sha256:d020037a86842d26d2b202f08688e2577c805b23b1c18db1a5c9bbaea3e72e93
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c1347d261e8d578be07dfc3db010c1c0747521c2443596e11a6fc584f3aa3e96
+oid sha256:cb798c7be90ed7deb3c91241b732db42360ec1053cf35da0121e50d76d308cde
 size 4040701744