Outer Step 2. Inner Step 64. Batch Size 24

config.json

@@ -268,18 +268,18 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5072860,
-    5072865,
-    5072869,
-    5072873,
-    5072878
+    5072896,
+    5072900,
+    5072904,
+    5072908,
+    5072913
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 59,
+  "inner_step": 64,
   "inner_steps": 0,
   "last_allreduce_block": 4016775,
   "layer_norm_epsilon": 1e-05,

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8818cddb505f46d1bd4aff6a8ae3b3fb207ec9e98d463bc3efd535b4087db216
+oid sha256:1a5a07b9016dd77b7b5a92ea60952b937635d915c0b4119d1a7843a5a283177f
 size 4040701744