Run 3. Outer Step 11. Inner Step 25.

config.json

@@ -268,18 +268,18 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5432100,
-    5432101,
-    5432103,
-    5432105,
-    5432106
+    5432130,
+    5432132,
+    5432133,
+    5432134,
+    5432136
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 23,
+  "inner_step": 25,
   "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 sha256:731d47f4a68f9f0334d03b873c3daba7ce2548acc72b721a65930a001f4ba38f
+oid sha256:4ddac83b3096703c9150efef5bef8cf034be2c02bfcabdea80f3b28a38a13d45
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:88cf88930e00fdd5984b65f8ffa740435ee1abcd8a33aa28399dd68b12e91c44
+oid sha256:78d4a8c5a3c0f2e8cb52d95c8b8ba72af0f2a8f7969efd7861573660544c1337
 size 4040701744