Run 4. Outer Step 24. Inner Step 31.

config.json

@@ -268,18 +268,24 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5532474,
-    5532478,
-    5532480,
-    5532482,
-    5532485
+    5532487,
+    5532489,
+    5532491,
+    5532493,
+    5532495,
+    5532497,
+    5532500,
+    5532501,
+    5532504,
+    5532506,
+    5532509
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 30,
+  "inner_step": 31,
   "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 sha256:666187cbcb7b48171bce5decc381edf34f5272a11341290f4a897830fe5b0ebd
+oid sha256:65238d00ab1ecb9e8e89756c2ce1048f8f92e6ff03e5a6b539ed55deab1fbb86
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:040492249383211d742e465bad70207e3ac024e43bfddeafe5a97a75172558dd
+oid sha256:3bb0e6f7fa6ae1712dff045c3306f29b8984cac8414fc037763749b87b1396be
 size 4040701744