Outer Step 10. Inner Step 38. Batch Size 348

config.json

@@ -268,23 +268,23 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5319559,
-    5319561,
-    5319564,
-    5319566,
-    5319568,
-    5319570,
-    5319571,
-    5319574,
-    5319576,
-    5319579
+    5319607,
+    5319609,
+    5319611,
+    5319613,
+    5319616,
+    5319619,
+    5319621,
+    5319623,
+    5319624,
+    5319626
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 31,
+  "inner_step": 38,
   "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:eda253a4e8c009cd3365e9324766ebba6d763899c7b1c7d5a9c7a1fb2a4f9512
+oid sha256:b0b3b394ff24d71cd40643eb80635bbfabfdca4fe64053ca33f97f7e25258462
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:88f7e51e70d173c04ab8fbba66cdafc17659cf1bc884cc5b33c803af90b8729e
+oid sha256:ef0722dd34ba4bfd728eefc76b76727546b7df55f4f9b355cc679ffecc3fa5be
 size 4040701744