Run 4. Outer Step 33. Inner Step 21.

config.json

@@ -268,18 +268,18 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5545312,
-    5545317,
-    5545321,
-    5545326,
-    5545330
+    5545335,
+    5545340,
+    5545345,
+    5545350,
+    5545354
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 20,
+  "inner_step": 21,
   "inner_steps": 0,
   "last_allreduce_block": 5543976,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e3c599ea6e9ea03a2ca4bb3a20fe65f1f5721a617fb72937e100589434517667
+oid sha256:6e3821a320822ce02440ff16c49b6148fa8e888e2731513a6f97924b27c341ef
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9efe024850412e8bee642db3a67f48a464cda26f4e923290de81962c63ea7b23
+oid sha256:ef002be3c24c91cf355f674f3b11f59bd87d7becdb50452f26e06d937e21fe2b
 size 4040701744