Run 4. Outer Step 29. Inner Step 31.

config.json

@@ -268,20 +268,18 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5539478,
-    5539482,
-    5539485,
-    5539488,
-    5539491,
-    5539493,
-    5539496
+    5539499,
+    5539501,
+    5539503,
+    5539505,
+    5539508
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 30,
+  "inner_step": 31,
   "inner_steps": 0,
   "last_allreduce_block": 5538408,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:912ac3dddb3ac65b3522cd099b3190d7d2819a25ab9b2ca62a43210d8bb92fe8
+oid sha256:f9a3425fabf00c60fd3d268e614cd4bd85f70d2ee3af0ca062c3626aa8873065
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7bed994e6ac83d1b3065172c690c7cda30d6a044b44520ad1d9b0ad0b3e1d2ba
+oid sha256:6c6ec226a2b015208b9d71f750c404ce0eb9cf612ee9a1297eb5af5c610d9959
 size 4040701744