Run 4. Outer Step 23. Inner Step 22.

config.json

@@ -268,18 +268,24 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5531546,
-    5531548,
-    5531549,
-    5531552,
-    5531554
+    5531556,
+    5531558,
+    5531560,
+    5531562,
+    5531565,
+    5531566,
+    5531568,
+    5531570,
+    5531572,
+    5531574,
+    5531576
   ],
   "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": 5529444,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c14c0f7c594b47f4d16be671032fb8d23f9426b45117aa8e33371d255cd2a30c
+oid sha256:13bd911bbe41366872ff2dc7d8f97fff2bff632064a41d9a6f042ba1c504bf6b
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:696263f1e6b3bbb7f4c54103d5ae1aeb57f0c0d0154c7ddee4c3eb1491606411
+oid sha256:0363fbd452715eed3ecf0d6a3ae674a7a0972300b184190f79cb385ff7ded6d7
 size 4040701744