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": [
-    5543803,
-    5543806,
-    5543808,
-    5543809,
-    5543810,
-    5543812,
-    5543814
+    5543817,
+    5543820,
+    5543823,
+    5543825,
+    5543827
   ],
   "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:b881866c0cb8c6ca32ccaae6b2c4f7efaca3c425f64661c824fa8d331806a3ac
+oid sha256:de5387bbab3c1d16585725fa50e90927c6f3ce526744489da92b273fefea00a5
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:07908b3189114703fa6289463a3103a9073cc7a3f85cea258c5f74db12766004
+oid sha256:44ad44a68eb5a4a26c4c72034ba51cc8345743d14c1c8d491c1ef08618b7a485
 size 4040701744