Run 4. Outer Step 20. Inner Step 16.

config.json

@@ -268,34 +268,29 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5525150,
-    5525151,
-    5525152,
-    5525153,
-    5525155,
-    5525157,
-    5525158,
-    5525160,
-    5525161,
-    5525163,
-    5525164,
-    5525166,
-    5525168,
-    5525170,
-    5525172,
-    5525174,
-    5525176,
-    5525178,
-    5525180,
-    5525182,
-    5525183
+    5525185,
+    5525186,
+    5525188,
+    5525190,
+    5525192,
+    5525194,
+    5525195,
+    5525195,
+    5525196,
+    5525196,
+    5525196,
+    5525197,
+    5525198,
+    5525200,
+    5525201,
+    5525206
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 15,
+  "inner_step": 16,
   "inner_steps": 0,
   "last_allreduce_block": 5523943,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9e3ce3e60c0d90b33ded929ecb91235ff48622e0dae60af8556654d6c0626a29
+oid sha256:a2d4fa858bbfcea4e19a7d0760df6b689c363c03270a651aab2ee2333176018d
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:70cc9a14fad97aa6c0d56bb707ecfa0651762d1fe6d0d73f626afaf9925127b1
+oid sha256:9b6be164ce9fa02f14eafc6a0a9fa8af10631bd0052033aac1ecabb5144802b2
 size 4040701744