Run 4. Outer Step 4. Inner Step 38.

config.json

@@ -268,20 +268,18 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5487540,
-    5487543,
-    5487546,
-    5487548,
-    5487550,
-    5487553,
-    5487556
+    5487559,
+    5487562,
+    5487564,
+    5487566,
+    5487569
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 37,
+  "inner_step": 38,
   "inner_steps": 0,
   "last_allreduce_block": 5485540,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9b3607816888585823f7adbef5afca4f53be33afa0664c965ddaa443c77656b8
+oid sha256:f30dd0342d9adf39d3808abe32334f86f9f40f5b4455b62af277103f10a65fd7
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1dda6b051dcba559849941f68c0f74561757ac9a77f843d342fc0947c3123ad4
+oid sha256:a471f28a9342bfd2b300e19893eb09bbd4e81910a9335d4b73b3eccfb697488e
 size 4040701744