Run 4. Outer Step 33. Inner Step 23.

config.json

@@ -268,21 +268,18 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5544778,
-    5544781,
-    5544784,
-    5544787,
-    5544790,
-    5544793,
-    5544795,
-    5544798
+    5544800,
+    5544804,
+    5544805,
+    5544809,
+    5544811
   ],
   "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": 5543976,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6b9992e15d522e5512d47e1be6d9cb32881295ab4f5f1d27504ac26a54112e3a
+oid sha256:d62259c553d04fb1211cc5d3b9c5d26eb7fb9dbaa520b26500425a6f66c54deb
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:35b3707cb0eb76f9df330a26881339f5c201449d5ff15b1f6b39db36a1e38a20
+oid sha256:5529f29ad2931afe30da98a2c3cb3a63001c9432cda0e3af15d1648656753e5a
 size 4040701744