Run 4. Outer Step 24. Inner Step 36.

config.json

@@ -268,24 +268,18 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5532617,
-    5532619,
-    5532621,
-    5532623,
-    5532625,
-    5532628,
-    5532630,
-    5532632,
-    5532633,
-    5532635,
-    5532638
+    5532640,
+    5532642,
+    5532644,
+    5532646,
+    5532648
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 35,
+  "inner_step": 36,
   "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 sha256:008066fae94517468dc1aeab28b3a2fadc6d54f31eab8e9d08654461511ffa5c
+oid sha256:814e932da366be829acb8ccf7a1f438582168752bde5c0a29be2201e9ea8c52e
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f7c46a768b3024e8861aa450125c837f22ee264a5af6910a4709061f8228076d
+oid sha256:297b8b2afd82f0ed6f3e467b8d22cea1588304e14b05b1675b178682ee680ad8
 size 4040701744