Run 4. Outer Step 23. Inner Step 23.

config.json

@@ -268,18 +268,23 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5531567,
-    5531569,
-    5531572,
-    5531575,
-    5531577
+    5531580,
+    5531582,
+    5531585,
+    5531588,
+    5531590,
+    5531591,
+    5531594,
+    5531596,
+    5531598,
+    5531600
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 22,
+  "inner_step": 23,
   "inner_steps": 0,
   "last_allreduce_block": 5529444,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5771fd864f9410cd4195681cbe0c5c4d477db019df016edeeb28088bf90fbc6e
+oid sha256:7c331934d9e445dd3bd13e76b7931add5382ed0238dc8258c8f0fb7f05d59829
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:35dd6c3271c200931af2d5c84c2f3c4a655be3f79063215945083c42f5a4ece4
+oid sha256:4eb6050a2b9515385f604871fc3d2815544d704c899e840511ddc340e0202fab
 size 4040701744