Outer Step 0. Inner Step 28. Batch Size 196

config.json

@@ -268,18 +268,18 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5129860,
-    5129867,
-    5129873,
-    5129879,
-    5129885
+    5129906,
+    5129914,
+    5129929,
+    5129935,
+    5129947
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 24,
+  "inner_step": 28,
   "inner_steps": 0,
   "last_allreduce_block": 5072405,
   "layer_norm_epsilon": 1e-05,

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8650177e305267d4d3e3f548d0b5a7ef126372e4a1de7d185af5864ef7e97230
+oid sha256:b07d9a789363586928535f3aab6351ff94084687604526ce2136b037dcf50dce
 size 4040701744