Outer Step 10. Inner Step 45. Batch Size 508

config.json

@@ -268,23 +268,23 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5319603,
-    5319605,
-    5319608,
-    5319611,
-    5319614,
-    5319617,
-    5319619,
-    5319621,
-    5319623,
-    5319625
+    5319654,
+    5319658,
+    5319660,
+    5319663,
+    5319665,
+    5319667,
+    5319670,
+    5319673,
+    5319675,
+    5319677
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 38,
+  "inner_step": 45,
   "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 sha256:14b0884f3fa31372e6df95f965260e7f20148689bfe2ee09be95497478acc656
+oid sha256:e51c749db1f9daef0d41c2066f54b308a4bc080ebf1c83995da3592a0eb2b483
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2b995ee76168d4bf9e477706248fd4964bb97542a59ef324babb25e0bf837264
+oid sha256:125196361056f54ae5b1aa6244a18aee1419c09e6eb8dce7d49f93fb59bdc8ec
 size 4040701744