Run 4. Outer Step 50. Inner Step 38.

config.json

@@ -268,18 +268,19 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5572643,
-    5572648,
-    5572652,
-    5572656,
-    5572660
+    5572665,
+    5572669,
+    5572674,
+    5572678,
+    5572682,
+    5572686
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 37,
+  "inner_step": 38,
   "inner_steps": 0,
   "last_allreduce_block": 5556551,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9c2a30dca3376977f403db31b40532b22b1fdd2e0ae89717c68f314932094546
+oid sha256:3171d5c2968076cd1310717e948f1d0cd7883612550d1eadb0ff4dbf178b3cee
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e82ff3cdb7ef80d11091d03d587fe33969159002b328125c64872286837c65bd
+oid sha256:a9ed5d2135049339fd4d29ce1641b855bfec7975ee4329ff58ff2d829e789233
 size 4040701744