Run 3. Outer Step 6. Inner Step 200.

config.json

@@ -268,18 +268,18 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5388299,
-    5388302,
-    5388306,
-    5388310,
-    5388313
+    5388346,
+    5388350,
+    5388353,
+    5388357,
+    5388360
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 197,
+  "inner_step": 200,
   "inner_steps": 0,
   "last_allreduce_block": 5378990,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5f1755d68077814b06bad02b2937d8f2565afbb695e6094338b5021887364c82
+oid sha256:7f057b9d4d32aa8eb3c3e12682dfe9b0b806f6929787991483ea81eb10159cf4
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8c21d0a6854ca1dcee4d7130eb4f06558bb341a4988a6c8b3acc9c04eb95ae70
+oid sha256:b9fe5a6786d35e22a41256296fc84c4d9717a31f8236f9101d5db2aede5c95d3
 size 4040701744