Run 4. Outer Step 35. Inner Step 37.

config.json

@@ -268,18 +268,18 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5549334,
-    5549338,
-    5549342,
-    5549346,
-    5549350
+    5549355,
+    5549359,
+    5549363,
+    5549367,
+    5549371
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 36,
+  "inner_step": 37,
   "inner_steps": 0,
   "last_allreduce_block": 5548219,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6e48f8697d97e6281e58a5a5eaf62881998e9f00a854a328420558f22f86a61e
+oid sha256:c2abb86db7535ae333e623b10f4c07e76404908ca8f9578a2bb927518af8f1ef
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:35a8f822d05ad9a300c0d2d5c70a4dfe9f5a6411550dec35f99d1e65348e93b5
+oid sha256:5868ea4f47242786de977401a09c4bcfeee29428de30d185dcf9db125919bcfc
 size 4040701744