Run 4. Outer Step 32. Inner Step 33.

config.json

@@ -268,19 +268,18 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5543747,
-    5543750,
-    5543753,
-    5543756,
-    5543758,
-    5543761
+    5543764,
+    5543767,
+    5543770,
+    5543773,
+    5543776
   ],
   "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 sha256:0ff134d9e968b7d5ff987314db997d98ae5bd35bca1f4bf99364083ab6b76d7d
+oid sha256:23d35f419a3afa2099ad2c02bf52e07258eb19f795ade75c5917bf401ea113a8
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4b43db5452c402ce1f238e05db718e097b00cb4d09bd422471090d38a59e4745
+oid sha256:8c1f668e69edd7f95d61339cf776c8b782664541abaa696ecca215fbab1b3783
 size 4040701744