Outer Step 8. Inner Step 55. Batch Size 472

config.json

@@ -268,23 +268,23 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5291639,
-    5291642,
-    5291646,
-    5291649,
-    5291652,
-    5291656,
-    5291659,
-    5291663,
-    5291667,
-    5291670
+    5291709,
+    5291713,
+    5291716,
+    5291719,
+    5291723,
+    5291727,
+    5291730,
+    5291734,
+    5291737,
+    5291741
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 45,
+  "inner_step": 55,
   "inner_steps": 0,
   "last_allreduce_block": 5283755,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8b9a5158131866cd6d256fe2d901c100917df653a66be6aaf87c7f30fa554ec2
+oid sha256:5c6e409dbe384af6fadb99103b9f66eb5152c1649ddf20cd22c0ca042f30cc68
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8bebb49fb079a9dde80b02081ef9b46748780b7a68fccf05752ad8f7fcb94030
+oid sha256:6723b689ea74009eaedc1e3b52a5fe1576bcab1ee6529b7bad216e4a55633c46
 size 4040701744