Run 4. Outer Step 7. Inner Step 64.

config.json

@@ -268,18 +268,18 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5494654,
-    5494659,
-    5494665,
-    5494670,
-    5494675
+    5494681,
+    5494686,
+    5494691,
+    5494696,
+    5494702
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 63,
+  "inner_step": 64,
   "inner_steps": 0,
   "last_allreduce_block": 5492224,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a526c4fc6a45dd6a4af3007021bf03e23348131547a9cda8376636609cc08c89
+oid sha256:e029ef9ea77a09e5d1297e08ce1e158df6c66275af51f465b9e4754fbb5d7f70
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ddc041873a3df666a0f2de19ced259c37e23e73f1eef1c20797279f3330b6e88
+oid sha256:6a095ecd373fd7d309feec0b109cb113bf21c7372b04a08516fba06b0393b2ad
 size 4040701744