Run 4. Outer Step 32. Inner Step 31.

config.json

@@ -268,21 +268,18 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5543669,
-    5543671,
-    5543672,
-    5543672,
-    5543673,
-    5543675,
-    5543678,
-    5543681
+    5543684,
+    5543687,
+    5543690,
+    5543691,
+    5543692
   ],
   "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": 5542700,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5725b25351aadc22a9fdd5dc1555999048ea300c5993beaa307784e098d24623
+oid sha256:3064b3c3e96d107e4afe1b8cf54a0601902be41c1b0eba9bb27a52879b02fef9
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9a61203815f0a4d6dc608b639407fd3221deef5f89a14389373e9c29c04b838c
+oid sha256:3a81d6b174321f06b3a6f7be13c0651e09bf9a8ddba748f18535b36522ae217b
 size 4040701744