Outer Step 3. Inner Step 17. Batch Size 256

config.json

@@ -268,18 +268,18 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5076100,
-    5076104,
-    5076108,
-    5076111,
-    5076116
+    5076132,
+    5076136,
+    5076140,
+    5076144,
+    5076148
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 12,
+  "inner_step": 17,
   "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:b16898bc02673361fbee7f628c841c98e424bc423a8a254d2cd974dd4baad9a0
+oid sha256:9ce6a69da1a3e31cc7df34f4cf1b25f526ab36a57fc25f8db3f54aab485897e2
 size 4040701744