Run 3. Outer Step 9. Inner Step 100.

config.json

@@ -268,18 +268,18 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5410304,
-    5410308,
-    5410312,
-    5410316,
-    5410320
+    5410352,
+    5410357,
+    5410361,
+    5410365,
+    5410368
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 98,
+  "inner_step": 100,
   "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 sha256:4e940a194dfae29346620b47011c05686cca663c4b0ef4d3692ce2bf8294152f
+oid sha256:c5649ace0e47d7fec3ee601ae0fd2a723f076a8bd0010716e3aa91cd675eaddf
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:54be17d4c291885f75a8d2b4ea5b490b6e9886e72a08330383f67891194ef80b
+oid sha256:553117121456d773c106201e7a6c01d76396ae7395646cfb0468d7b41b4607fe
 size 4040701744