Run 4. Outer Step 33. Inner Step 22.

config.json

@@ -268,20 +268,18 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5544675,
-    5544677,
-    5544680,
-    5544683,
-    5544685,
-    5544687,
-    5544690
+    5544693,
+    5544696,
+    5544698,
+    5544700,
+    5544702
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 21,
+  "inner_step": 22,
   "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 sha256:786e6a003ab8a6c338d232871ea7e68809f6e03e46983c9ed6be4ce1b92edc16
+oid sha256:add560496d5494d4f637c924cf9d6bad4ba26b4e4797bbe907e36f9c4a5c4d4f
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d18cb3443206dffd804a8c7e1c420e6706ec7f70942e965dae236228f9ed2066
+oid sha256:600c502c5997779df246f21c69f209dd4c3f2e284db33e81fe38df823d8085ac
 size 4040701744