Run 4. Outer Step 28. Inner Step 21.

config.json

@@ -268,18 +268,24 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5537707,
-    5537709,
-    5537712,
-    5537714,
-    5537717
+    5537720,
+    5537722,
+    5537724,
+    5537726,
+    5537728,
+    5537731,
+    5537733,
+    5537735,
+    5537738,
+    5537740,
+    5537742
   ],
   "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": 5537093,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6a1f4bbc643ce51d1455dfb4564405f59b44531feaa7efe875777bc27ac605f3
+oid sha256:2d2b0a214fadad9a20a51b3f7040999719a1d06c375f49507dec66e76268eef6
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:611c442fa978cdba9be9abba0048b1ccf0881ce7c8bd5b31a8f2ecbf6f39ad6e
+oid sha256:90e1c32b395d50590329a71d689e51dadfeb40d49edcd9cca2acabb15f038047
 size 4040701744