Run 3. Outer Step 1. Inner Step 9.

config.json

@@ -268,18 +268,18 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5366453,
-    5366459,
-    5366466,
-    5366472,
-    5366479
+    5366507,
+    5366513,
+    5366519,
+    5366526,
+    5366532
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 7,
+  "inner_step": 9,
   "inner_steps": 0,
   "last_allreduce_block": 5366095,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7e7a0683b1233fac154aaf4128824df81962a21031778d2a8b2fff0ac3000d0d
+oid sha256:6c10ca02c7109cca1329cbadf7927829472674ba9ecdff6f72d88d229950713c
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d712233eeb41bed57f458e7dec41c79743712c318e6ed5ac8e6d5492f6496c18
+oid sha256:7dd7842b08e591d0e98760458bc8bf5aaac1cccd9f8269e436b4419674f4cbda
 size 4040701744