Run 4. Outer Step 33. Inner Step 4.

config.json

@@ -268,20 +268,18 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5544120,
-    5544123,
-    5544126,
-    5544127,
-    5544129,
-    5544132,
-    5544134
+    5544138,
+    5544140,
+    5544141,
+    5544143,
+    5544146
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 3,
+  "inner_step": 4,
   "inner_steps": 0,
   "last_allreduce_block": 5543976,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e670d147436e2960cc641d8a1b77bf0f3f67de89a0cd02e7a551630710eea5c1
+oid sha256:1e94d23119ed07cc3a74277c161a6229c67f01a978a8f55b0b04c19d8cc21cb1
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b6b45aac3bf4246c16bd3d16a7f720cb8d1da5cb82215d933c2e648eb2d76e5c
+oid sha256:45e39639f86ec95f755aac53a29a00a5cc8c5000cd57e7ee1494c6dd46a3ca78
 size 4040701744