Run 4. Outer Step 24. Inner Step 23.

config.json

@@ -268,24 +268,23 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5532227,
-    5532229,
-    5532231,
-    5532233,
-    5532235,
-    5532237,
-    5532239,
-    5532242,
-    5532244,
-    5532246,
-    5532249
+    5532251,
+    5532254,
+    5532257,
+    5532260,
+    5532262,
+    5532265,
+    5532267,
+    5532269,
+    5532271,
+    5532274
   ],
   "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": 5529444,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e6049aee1dc8bcbe595d040515c92a3ea070ea28efaff39aba8fbe13e0864565
+oid sha256:1124d805e1205741e85ef16bf0e2dffbb7e70dd5346b60661550c3227fbb5127
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9cafa7071937a1727a64a92995610dd94085b9820e346b752bda6f1bd0b42efa
+oid sha256:451c66b83e383176f64bb6604aa873725390a7d397b8b7117893f42c81bfcb40
 size 4040701744