Run 3. Outer Step 11. Inner Step 16.

config.json

@@ -268,18 +268,18 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5431939,
-    5431943,
-    5431945,
-    5431947,
-    5431949
+    5431976,
+    5431979,
+    5431979,
+    5431981,
+    5431982
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 14,
+  "inner_step": 16,
   "inner_steps": 0,
   "last_allreduce_block": 5422706,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7c4c06dc7bb1c66c501dd65a020c6bdf5185b1200c96531a9e009c4c923ae9b9
+oid sha256:64cc5ab9d9b6798f52951e53f7fdd43277ad4e2a427e64a242a6d5443f21fa7a
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a8c76a6a6dcae7aa26e5ac5407308b2f1d4527d210e94f5dc74ced4de8ddb489
+oid sha256:f5bc5bf3e4655d8bfed00da108bc4528acc5652da31ec81253ec58dda191b9ff
 size 4040701744