Outer Step 0. Inner Step 61. Batch Size 156

config.json

@@ -268,23 +268,23 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5139862,
-    5139867,
-    5139874,
-    5139881,
-    5139890,
-    5139901,
-    5139915,
-    5139922,
-    5139931,
-    5139937
+    5139976,
+    5139983,
+    5139990,
+    5140000,
+    5140007,
+    5140014,
+    5140020,
+    5140024,
+    5140030,
+    5140035
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 53,
+  "inner_step": 61,
   "inner_steps": 0,
   "last_allreduce_block": 5072405,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:fc83ae9faa37646faef447b36f6cf00be7961d1e2c1fe0689f549bb03e8df9f3
+oid sha256:24df5b2376c11779644d2afcf252fdf0b83ce3e2ef7c7bdb7584d0c10def5123
 size 8081781770

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ec42742096ce22788aee7ad317faf5e9d1e3a61743dad5587089d3e5db889828
+oid sha256:006e33b9842091bedc7e3c9e1d84beb1afc7d32a81d5b98ec0754f9e210c099e
 size 4040701744