Run 4. Outer Step 29. Inner Step 21.

config.json

@@ -268,18 +268,26 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5539162,
-    5539164,
-    5539167,
-    5539169,
-    5539172
+    5539175,
+    5539176,
+    5539177,
+    5539179,
+    5539180,
+    5539182,
+    5539184,
+    5539186,
+    5539188,
+    5539190,
+    5539192,
+    5539195,
+    5539198
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 20,
+  "inner_step": 21,
   "inner_steps": 0,
   "last_allreduce_block": 5538408,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:48a1f2b986092623ca4c492e995f5b2c6818351e90126d13d6a2a93ebfc84a05
+oid sha256:55a39609127656ef574f398a13eb2e31f5b25533555ba82c3d737b00f8e5b54e
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c832bb6ea2e28b8c0639935ad1a251fa08ae6322e5f0a3570d939ce15c4d75cf
+oid sha256:1556ccaddcfd0ea54664062eb2c2f3e19409067b33d9c20fc0edd96b1da8d844
 size 4040701744