Run 4. Outer Step 4. Inner Step 8.

config.json

@@ -268,20 +268,18 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5486603,
-    5486606,
-    5486609,
-    5486612,
-    5486614,
-    5486617,
-    5486619
+    5486623,
+    5486626,
+    5486629,
+    5486631,
+    5486635
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 7,
+  "inner_step": 8,
   "inner_steps": 0,
   "last_allreduce_block": 5485540,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a8676a03b591ffc49fc57641221d585120406652914d7536bcd9913eb6d2263e
+oid sha256:a5dc5cd414549c26ac227cad0a6630502745d81d51e6ac8d54779bec61e0cf08
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1d76ec3f18b1992bc6b6bfaa11cf42f47719b03171c3ef010b7b1bf33000e8e0
+oid sha256:01c4aa3b9c575ba673e85ffa61241eb0139a56e5868846a6c291b03434a5d9ff
 size 4040701744