Outer Step 10. Inner Step 45. Batch Size 8

config.json

@@ -268,23 +268,23 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5319609,
-    5319611,
-    5319613,
-    5319616,
-    5319619,
-    5319621,
-    5319623,
-    5319625,
-    5319627,
-    5319629
+    5319660,
+    5319663,
+    5319665,
+    5319667,
+    5319669,
+    5319669,
+    5319670,
+    5319672,
+    5319674,
+    5319676
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 38,
+  "inner_step": 45,
   "inner_steps": 0,
   "last_allreduce_block": 5297756,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1fc4639fb9d3cabc00aee635b7ad42545d43c9787d68e44fdf100e5124bece0b
+oid sha256:c78a06ef8f49b757b143857e1e993cdbfe9917bd68aa752b09a9b599561a2d8d
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e9b6f5967784ec750507401fd214a831b57c1bfe0a3b6703318fcf43e14e1fd9
+oid sha256:c7737983b4b1efba4b62b9c337feec2dc2d15a1ec3ad3e3d2609b08ead13c1d2
 size 4040701744