Run 4. Outer Step 4. Inner Step 20.

config.json

@@ -268,21 +268,18 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5486976,
-    5486978,
-    5486981,
-    5486984,
-    5486987,
-    5486989,
-    5486991,
-    5486993
+    5486996,
+    5486998,
+    5487000,
+    5487003,
+    5487006
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 19,
+  "inner_step": 20,
   "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 sha256:8fd480c6e187f18f788567a8c5157d2dbb14a30100627ef3127ab538b67c5bed
+oid sha256:71ca66a58e64b8274160a0212418cbabba62fbda993598383bd898c9cc296d09
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:453b486a6435aac5b1ab2fafe8a6ef7782346061530f04898190dbda9cc3bb3e
+oid sha256:8d0ff5a6fa0134327819764d4d73083aef0ce655b5955e51d3381b9e0939226a
 size 4040701744