Run 4. Outer Step 32. Inner Step 32.

config.json

@@ -268,18 +268,18 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5543712,
-    5543714,
-    5543716,
-    5543719,
-    5543721
+    5543723,
+    5543726,
+    5543729,
+    5543732,
+    5543734
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 31,
+  "inner_step": 32,
   "inner_steps": 0,
   "last_allreduce_block": 5542700,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3064b3c3e96d107e4afe1b8cf54a0601902be41c1b0eba9bb27a52879b02fef9
+oid sha256:451516643a14dd8bd4004a1e8e8098d4b44599c17800e3259c62d6743702a5b9
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3a81d6b174321f06b3a6f7be13c0651e09bf9a8ddba748f18535b36522ae217b
+oid sha256:49c3f11779fe91a37edb0ab2e4786bbaff4d7c34393e690a2620be549da4ef20
 size 4040701744