Run 3. Outer Step 11. Inner Step 256.

config.json

@@ -268,18 +268,18 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5444643,
-    5444647,
-    5444650,
-    5444654,
-    5444658
+    5444686,
+    5444690,
+    5444693,
+    5444697,
+    5444701
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 254,
+  "inner_step": 256,
   "inner_steps": 0,
   "last_allreduce_block": 5422706,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b50018b5202792256f9f40133e250ca935a3e5712a1412049a11b7707298d9f0
+oid sha256:3e7cfb15bf46f07368f68bbda4012fd2f7b2e286988aad19154a35bf4cfeae43
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5634627a0ded3b492e7191f91c5f89280a9c17662b46b16ed2c22540ba7100e4
+oid sha256:7ba2c550983d5943b499fba14c310624c9467e2f5b3a517a21568d74387a2494
 size 4040701744