Run 4. Outer Step 66. Inner Step 101.

config.json

@@ -268,18 +268,18 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5621999,
-    5622003,
-    5622007,
-    5622011,
-    5622015
+    5622020,
+    5622024,
+    5622028,
+    5622032,
+    5622036
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 100,
+  "inner_step": 101,
   "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 sha256:843547cf935d72139be946bfa04f4d2ab176fdc3ea9b5289fd7605b32446ebd9
+oid sha256:9481ed1fde3d1511fc5c6bc17ee905e418d85fdc26dfe0a077a896a0a711d7f5
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5312e03576728754baac14d767d09886dc2f8d0e8b076bef6976680e69ed739b
+oid sha256:0c42c9d9bf40129f8c20ef25e31be0c2232c4952bdb8599c7fb8a55f0b18855c
 size 4040701744