Run 4. Outer Step 20. Inner Step 21.

config.json

@@ -268,30 +268,30 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5525255,
-    5525256,
-    5525257,
-    5525259,
-    5525260,
-    5525262,
-    5525263,
-    5525264,
-    5525265,
-    5525265,
-    5525266,
-    5525268,
-    5525269,
-    5525271,
-    5525273,
-    5525275,
-    5525277
+    5525278,
+    5525280,
+    5525282,
+    5525283,
+    5525284,
+    5525285,
+    5525287,
+    5525288,
+    5525290,
+    5525292,
+    5525293,
+    5525294,
+    5525295,
+    5525296,
+    5525297,
+    5525299,
+    5525300
   ],
   "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": 5523943,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:fa6f20d28b5315eaacb084c94b6cf4b182216a929b7d7f879c1352a09b89204a
+oid sha256:76ea6d14fd83f4d847955188ed8bb1da758301e01b86fa57dd92fc5c1f4711d5
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c8725f87fcff2c2b111c87f97b63dd0b68e8927e19890564c98aa9594155e0a2
+oid sha256:151e2d48eb1057448e578963f9d4b6e328bdfbe3a4c08920d5a8af0f485200f3
 size 4040701744