Outer Step 0. Inner Step 139. Batch Size 488

config.json

@@ -268,23 +268,23 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5155901,
-    5155904,
-    5155909,
-    5155914,
-    5155918,
-    5155922,
-    5155926,
-    5155930,
-    5155935,
-    5155939
+    5155971,
+    5155978,
+    5155983,
+    5155991,
+    5155998,
+    5156012,
+    5156021,
+    5156028,
+    5156034,
+    5156041
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 131,
+  "inner_step": 139,
   "inner_steps": 0,
   "last_allreduce_block": 4016775,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:82dc24b45797cc41dcad8a328a6c86daf05c96e238ae1fa05b6abfdc2eb23f07
+oid sha256:29d8505b18315249daf98332e78fc5d214085c934cef502a29f3485b85dfd45f
 size 8081781770

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c58c6cc4ad6b89d00c4258037daadd573379735216c1da25bbcc880ed7f69cfb
+oid sha256:e0cf81c44eb1fa355c77ebae7ab452b68c30430a7cc380fa492268ede68e3fea
 size 4040701744