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": [
-    5543811,
-    5543814,
-    5543816,
-    5543818,
-    5543821,
-    5543823,
-    5543825,
-    5543827,
-    5543829
+    5543832,
+    5543834,
+    5543836,
+    5543838,
+    5543841
   ],
   "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 sha256:04af216c4becf8ecb933c258ed246f10115fdce4929593f00624534433449e92
+oid sha256:c7f4d1ddb10718f36b1e946e0101431e17bdc4dc7af30eb5283f0a75904c270c
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a5533c78b13ce1f8896217e7ae07a66b74b32f0a1d680721974f91f79977974a
+oid sha256:f020f45db07eebf1ebcaa805190c8e6ec25c42bbfa36b7e5d4379322c7980799
 size 4040701744