Run 3. Outer Step 10. Inner Step 22.

config.json

@@ -268,18 +268,18 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5422445,
-    5422448,
-    5422451,
-    5422455,
-    5422458
+    5422490,
+    5422494,
+    5422497,
+    5422500,
+    5422503
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 20,
+  "inner_step": 22,
   "inner_steps": 0,
   "last_allreduce_block": 5398844,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:341bca7650b7083f70cb120ea0c8e8ec3a771325fb75a8082520dbab9f174bac
+oid sha256:9e8c46e3e86e633b4f9dfa5b3a1e809cbf70e54feaeb4f41f00a8dee7b8ee867
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c31e2765af33bd922cc82fe29f5856867b9f76a05f2b01bb758447836218a2ae
+oid sha256:5c91b02bf2af7ca089811ef38eadcd39e3c391f58eed8ac77d0dba709cf9f6b8
 size 4040701744