Run 4. Outer Step 33. Inner Step 22.

config.json

@@ -268,18 +268,26 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5544637,
-    5544641,
-    5544644,
-    5544646,
-    5544648
+    5544651,
+    5544654,
+    5544656,
+    5544658,
+    5544661,
+    5544663,
+    5544666,
+    5544668,
+    5544669,
+    5544669,
+    5544673,
+    5544675,
+    5544676
   ],
   "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:f8a45216aab717b870199babb79be129b5029cc25ccffe9dbca2df58bd0b8016
+oid sha256:be85640a184222f28a3cb79f4268436e6229cd2f61c6d8f7f64f00b24a9d4f9a
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a796625b50aa5f590a6c14060bfcdb1105e4e74b44556bc8d643d99dfd5683e1
+oid sha256:3d20b599fe55419e87724f8e26796e47faad98124b2da072379a2782a0706040
 size 4040701744