Run 4. Outer Step 32. Inner Step 38.

config.json

@@ -268,20 +268,18 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5543885,
-    5543888,
-    5543890,
-    5543892,
-    5543894,
-    5543896,
-    5543898
+    5543901,
+    5543904,
+    5543906,
+    5543909,
+    5543911
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 37,
+  "inner_step": 38,
   "inner_steps": 0,
   "last_allreduce_block": 5542700,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:278fa955207bb8c68779282663df3d7533c90a833aa21cc4bdd93a502f0234cb
+oid sha256:9dbf96ebd3d31d80fa5c47aa0b599d9e31568d811e8f34d2094beeb73829355c
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:48b1abc57605e8529b87b2ddb7010de841aeb44ab0497a9ecfea9a503e9bd583
+oid sha256:c9e599dad4cd2584b94ec0add60d05eae671d7af8eb7a322e999fac89bfb95a1
 size 4040701744