Run 4. Outer Step 20. Inner Step 18.

config.json

@@ -268,27 +268,30 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5525180,
-    5525182,
-    5525183,
-    5525184,
-    5525186,
-    5525188,
-    5525189,
-    5525190,
-    5525192,
-    5525193,
-    5525194,
-    5525196,
-    5525198,
-    5525200
+    5525202,
+    5525202,
+    5525204,
+    5525205,
+    5525206,
+    5525207,
+    5525208,
+    5525210,
+    5525211,
+    5525212,
+    5525213,
+    5525214,
+    5525215,
+    5525216,
+    5525218,
+    5525220,
+    5525222
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 17,
+  "inner_step": 18,
   "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:cf7cfdaed7389ba468324b6dd85ae3c9c970b187afc1cb7caa0c404dd96589f7
+oid sha256:04d17d5620181d48753998a00253cbd62bbd5ad1f25cafd036ab2d03bb273d58
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:bea61a6d18e76d13e992ddbd8cc5a9f1e186281bd327c17b4162f008f4e89011
+oid sha256:d374bd79a383532460827fff4e97b7afd64ec19e93671e7b358f1b5f72b1601c
 size 4040701744