Run 4. Outer Step 35. Inner Step 21.

config.json

@@ -268,18 +268,18 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5548757,
-    5548761,
-    5548765,
-    5548770,
-    5548774
+    5548778,
+    5548783,
+    5548787,
+    5548791,
+    5548796
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 20,
+  "inner_step": 21,
   "inner_steps": 0,
   "last_allreduce_block": 5543976,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5dbb0d755a16aea563156f23ee20424ab0a0a78170d2995ccd6d6b67f775a528
+oid sha256:9b8cdf73ee268c9efca46af074eaf506d12ee8568807a44201f6538f0a937e6f
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:63abf3213a9958c001ddabd583e53e91d96085a725743e90477bef5d78725586
+oid sha256:1bb6e18d4d67b656bcd41fe48f23c4feb77cf7e6707de453de719752f85aeac2
 size 4040701744