Run 4. Outer Step 24. Inner Step 37.

config.json

@@ -268,18 +268,23 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5532640,
-    5532642,
-    5532644,
-    5532646,
-    5532648
+    5532651,
+    5532655,
+    5532658,
+    5532661,
+    5532663,
+    5532665,
+    5532667,
+    5532668,
+    5532671,
+    5532672
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 36,
+  "inner_step": 37,
   "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 sha256:814e932da366be829acb8ccf7a1f438582168752bde5c0a29be2201e9ea8c52e
+oid sha256:c28d92d8ef1c53fba1b321a5d5557439b55d0eafb6cb2b5608628bc83fbce6bf
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:297b8b2afd82f0ed6f3e467b8d22cea1588304e14b05b1675b178682ee680ad8
+oid sha256:4df66243e2964fa0bd0be554a554195afa4821e69b8cee07f1c46ef1a2e845ea
 size 4040701744