Run 4. Outer Step 32. Inner Step 33.

config.json

@@ -268,22 +268,18 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5543737,
-    5543739,
-    5543741,
-    5543742,
-    5543744,
-    5543746,
-    5543748,
-    5543749,
-    5543751
+    5543754,
+    5543755,
+    5543759,
+    5543761,
+    5543763
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 32,
+  "inner_step": 33,
   "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 sha256:451516643a14dd8bd4004a1e8e8098d4b44599c17800e3259c62d6743702a5b9
+oid sha256:597c827600b050b037fb64ea258b0a446d83ff8e34b39f153fd7e9788520d5d0
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:49c3f11779fe91a37edb0ab2e4786bbaff4d7c34393e690a2620be549da4ef20
+oid sha256:c170de22be7ec6fe2db0b8f6951f2a6326ccb3919b7dc2f59824eb1b5d589702
 size 4040701744