Run 2. Outer Step 11. Inner Step 29.

config.json

@@ -268,18 +268,18 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5363841,
-    5363846,
-    5363850,
-    5363854,
-    5363859
+    5363868,
+    5363872,
+    5363876,
+    5363881,
+    5363885
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 28,
+  "inner_step": 29,
   "inner_steps": 0,
   "last_allreduce_block": 5358692,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2b7a10558ee87317b7c57fbed386d6ea54b6de4aa33b4d658795ac1680b158cf
+oid sha256:b7c1b64e2eb8fc6b177601bb11757a6f94341b14ac9ec3eb8df9cc2bd5a65013
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:fd40474ff653702bd228903cb07c3d53763c5c5e0b56d133ccfd3115a0bccd32
+oid sha256:09ff5a5b02a3c9b14e13d524e4b05e9c8c821f79a6c269bd6232886cd8f5dfb8
 size 4040701744