Run 2. Outer Step 3. Inner Step 34.

config.json

@@ -268,18 +268,18 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5330782,
-    5330786,
-    5330790,
-    5330794,
-    5330798
+    5330829,
+    5330833,
+    5330837,
+    5330841,
+    5330845
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 27,
+  "inner_step": 34,
   "inner_steps": 0,
   "last_allreduce_block": 5323221,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e17af5a4b51e83223f1e043596100e40bca7b5726262daa50e41d8f29f7e7344
+oid sha256:5af41e65f2e56a280fe7f803b7991c61ecdc99b4b14cdb11edf0af81bf84e236
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d7db3ed299810fb9b7a86ba9153b9a46b81ec2fdcb3eba8d137da2aed8b45893
+oid sha256:9170f6297ebe94fbf91de0c080c30372121204cdafc30437c2f2e45820018880
 size 4040701744