Outer Step 9. Inner Step 310. Batch Size 196

config.json

@@ -268,23 +268,23 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5318897,
-    5318899,
-    5318901,
-    5318903,
-    5318906,
-    5318908,
-    5318910,
-    5318913,
-    5318914,
-    5318916
+    5318943,
+    5318946,
+    5318948,
+    5318949,
+    5318951,
+    5318953,
+    5318955,
+    5318956,
+    5318956,
+    5318956
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 304,
+  "inner_step": 310,
   "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 sha256:468b7d22f5abcc52b655099e18f90d97bbd68a88add305447b5ddfdfe7355efd
+oid sha256:9e255a04a34e81713c922223d00b95053a4afcb36911d56cc5cc742d2ecf952f
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2a0ae8f91e5cc99699e8f26571cc117e21fc72a058ffd60895380ec97408e1e7
+oid sha256:f4096e68892b4c5a68ab0fd8b99099f088bced544fb7af71dfc4f0d06ef4dfdf
 size 4040701744