Run 4. Outer Step 22. Inner Step 3.

config.json

@@ -268,18 +268,18 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5526658,
-    5526664,
-    5526670,
-    5526677,
-    5526684
+    5526692,
+    5526699,
+    5526706,
+    5526712,
+    5526719
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 2,
+  "inner_step": 3,
   "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:cde7de2bbb6e3914446e9dceb22cab34d305e4b9abec58f498142f3fb51ce93b
+oid sha256:499924ebc7f3375a993ecbf44014b3b9e00053bfa661506b86e973e05572c2f8
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a4d1afd20a186f1df1ee96956fe853265acf4037aa11761fc2a565b4ecb87167
+oid sha256:f810ca517345784d90b2984d3f9267d48b5e5b9f2d2c95497cdb0e3f326b17a8
 size 4040701744