Outer Step 4. Inner Step 1078. Batch Size 416

config.json

@@ -268,23 +268,23 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5240012,
-    5240016,
-    5240020,
-    5240025,
-    5240029,
-    5240033,
-    5240038,
-    5240043,
-    5240047,
-    5240051
+    5240083,
+    5240087,
+    5240091,
+    5240095,
+    5240099,
+    5240104,
+    5240108,
+    5240113,
+    5240117,
+    5240123
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 1069,
+  "inner_step": 1078,
   "inner_steps": 0,
   "last_allreduce_block": 5226437,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:07841b45bb9e363af69e4c7127b06fecd985af58114f7bf7e73a0f0ebd4a81bd
+oid sha256:0fc1c27e97cd61c99cb42e8f5bf00c0ac28f0745a46e54593464c54434e3528a
 size 8081781770

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b633caa7ce689a35fa0acbf16dcdeebf29f3933a314e16140416de4837375ac0
+oid sha256:d9d7884014524456efa2f6decd9e04d732b6baa1d5c78925b7d44a4a20a32e0d
 size 4040701744