Run 4. Outer Step 30. Inner Step 12.

config.json

@@ -268,18 +268,18 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5540876,
-    5540881,
-    5540885,
-    5540889,
-    5540894
+    5540898,
+    5540902,
+    5540907,
+    5540911,
+    5540915
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 11,
+  "inner_step": 12,
   "inner_steps": 0,
   "last_allreduce_block": 5539810,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:42fc981006ceca97c2fc35042f14014fd6d2076e2319e883287f3e763e9963ba
+oid sha256:17890dea70c7d827ee0e85b531f77f3b9e77d206b28c95bf6a567af6c0c5e0a7
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8e2366bfd5eed22dfcd85f9a0d0c60db9b9ffca9ca9d04cb8ac5de3c4b9e87c9
+oid sha256:238bc4ef53fcc52e8e9ae81775cc6cbd98163453933faf10011523fe7da1a57b
 size 4040701744