Run 4. Outer Step 63. Inner Step 62.

config.json

@@ -268,9 +268,9 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5613876,
-    5613876,
-    5613876,
+    5613877,
+    5613877,
+    5613877,
     5613877,
     5613877
   ],
@@ -279,7 +279,7 @@
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 61,
+  "inner_step": 62,
   "inner_steps": 0,
   "last_allreduce_block": 5611658,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7af9ca09a832d00b7ef3983ceb8ddaa8865e3f08e3b3aaf1262aa62c0aa5f70a
+oid sha256:8b9716f3d5e95bc382adac1c53b8c31e8f2f078543c1562f0487b368d61b80d3
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2f363744e6766d92e3b1821fd2a3e970e7f2b7fcc778db9249d160846804851e
+oid sha256:e3f94f380f25aff7a3133ce274d47f36460ae33004d19a5d1cb44adaea11b0bf
 size 4040701744