Run 4. Outer Step 20. Inner Step 15.

config.json

@@ -268,30 +268,34 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5525125,
-    5525127,
-    5525129,
-    5525130,
-    5525132,
-    5525133,
-    5525134,
-    5525136,
-    5525137,
-    5525138,
-    5525139,
-    5525141,
-    5525142,
-    5525143,
-    5525144,
-    5525146,
-    5525147
+    5525150,
+    5525151,
+    5525152,
+    5525153,
+    5525155,
+    5525157,
+    5525158,
+    5525160,
+    5525161,
+    5525163,
+    5525164,
+    5525166,
+    5525168,
+    5525170,
+    5525172,
+    5525174,
+    5525176,
+    5525178,
+    5525180,
+    5525182,
+    5525183
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 14,
+  "inner_step": 15,
   "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:d68c442fca57dbf2cdcebcdb5b000fe110ffe252f020fe3ba9be0289c7f6a4bd
+oid sha256:9e3ce3e60c0d90b33ded929ecb91235ff48622e0dae60af8556654d6c0626a29
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:40842c7187b2d1b310a7844853a8652452ff713afb59ecbe8ba1e5e2c4cf7d8e
+oid sha256:70cc9a14fad97aa6c0d56bb707ecfa0651762d1fe6d0d73f626afaf9925127b1
 size 4040701744