Run 4. Outer Step 29. Inner Step 38.

config.json

@@ -268,18 +268,22 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5539664,
-    5539667,
-    5539669,
-    5539670,
-    5539673
+    5539676,
+    5539679,
+    5539681,
+    5539683,
+    5539685,
+    5539688,
+    5539691,
+    5539694,
+    5539696
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 37,
+  "inner_step": 38,
   "inner_steps": 0,
   "last_allreduce_block": 5538408,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:987c1a50f273fe733f005e73f4f286f8a8dc49bb8906ddd131e7f640fa178c16
+oid sha256:cbe050f993e8637b02c0443ee9cfabda2ace0580fe4dd8717043b703a16c15d8
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7fdc5f2ceae5b4d2537d25128999e6ef9f7f5ca34f675a3c4f1806b1c59316a9
+oid sha256:94cf8ddf0a931381dc5728f319efbe08eb2506fe6d07e893a8d19d7e1c1cb994
 size 4040701744