Outer Step 10. Inner Step 45. Batch Size 308

config.json

@@ -268,23 +268,23 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5319607,
-    5319609,
-    5319611,
-    5319613,
-    5319616,
-    5319619,
-    5319621,
-    5319623,
-    5319624,
-    5319626
+    5319656,
+    5319658,
+    5319661,
+    5319662,
+    5319665,
+    5319667,
+    5319669,
+    5319671,
+    5319674,
+    5319676
   ],
   "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:b0b3b394ff24d71cd40643eb80635bbfabfdca4fe64053ca33f97f7e25258462
+oid sha256:9c4bc093fa0e4ad769e65f944d0ac328f5991f543806cc35f0bcae638cb3353c
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ef0722dd34ba4bfd728eefc76b76727546b7df55f4f9b355cc679ffecc3fa5be
+oid sha256:e4c104e11b23004e7d5ff28efa1f7872f0cb79eb190c312adf604deded93195c
 size 4040701744