Run 3. Outer Step 9. Inner Step 1000.

config.json

@@ -268,18 +268,18 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5421144,
-    5421147,
-    5421152,
-    5421155,
-    5421158
+    5421185,
+    5421198,
+    5421201,
+    5421204,
+    5421208
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 998,
+  "inner_step": 1000,
   "inner_steps": 0,
   "last_allreduce_block": 5398844,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:87dee1281467083539b835a7c3c62e2f7ea2a099ad57620fffef61bd409a361c
+oid sha256:ba8903039b5398d46cb4920c5d62a61d86d9e6325134cb74cb02ddd17180ab79
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:97de0a9d30207e4bb75ae65f3c105f2b55bebdbf9fcd7a3fa74a3103006bdd9b
+oid sha256:fb191848407059ce3eea46a8e10237cf77cf6b0fc119f357b2a0f00dc1dd4e46
 size 4040701744