Run 4. Outer Step 66. Inner Step 32.

config.json

@@ -268,18 +268,20 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5620677,
-    5620679,
-    5620682,
-    5620685,
-    5620687
+    5620690,
+    5620692,
+    5620695,
+    5620696,
+    5620699,
+    5620702,
+    5620704
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 31,
+  "inner_step": 32,
   "inner_steps": 0,
   "last_allreduce_block": 5619221,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a149063878af0a058739726fea2fa5fa8b8267de8b4af3045e976e1c927dbe4f
+oid sha256:dcf460f643648c387766cc3e1e0f6e2f4b7ac256d0b8b995c4cf4ff1ac97275f
 size 8081782026