Run 4. Outer Step 3. Inner Step 133.

config.json

@@ -268,18 +268,23 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5484736,
-    5484740,
-    5484742,
-    5484745,
-    5484748
+    5484751,
+    5484753,
+    5484756,
+    5484759,
+    5484761,
+    5484763,
+    5484766,
+    5484770,
+    5484773,
+    5484775
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 132,
+  "inner_step": 133,
   "inner_steps": 0,
   "last_allreduce_block": 5477809,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:83b36b43c49536779fb00fc6f8b30d3637e0cc97b6b398bb5be3bdbd39ffab2e
+oid sha256:66a905a382c3b9f85305dfc6d2f6e84a418bd1c99bc0272efa375acb18627f13
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:944e1bb5ae986765eba4267e8325ee486605f675d39f23a803aa1108f600afb7
+oid sha256:55e3754d4ccdb05d15ddc4ee343200f6a769b683f6b62790ab1a357b6098a7f1
 size 4040701744