Run 4. Outer Step 4. Inner Step 7.

config.json

@@ -268,21 +268,18 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5486567,
-    5486570,
-    5486573,
-    5486577,
-    5486580,
-    5486581,
-    5486583,
-    5486585
+    5486589,
+    5486592,
+    5486594,
+    5486596,
+    5486600
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 6,
+  "inner_step": 7,
   "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:dd2dffed63f719a50efeaf65cf37ffd482417f2ade8b8ed4172d32097ae6e7f4
+oid sha256:a8676a03b591ffc49fc57641221d585120406652914d7536bcd9913eb6d2263e
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d2082ea641862ca8ad95440fe3338a39a44f1f093c5c27d4fc97d7ae11904250
+oid sha256:1d76ec3f18b1992bc6b6bfaa11cf42f47719b03171c3ef010b7b1bf33000e8e0
 size 4040701744