Run 3. Outer Step 6. Inner Step 298.

config.json

@@ -268,18 +268,18 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5389928,
-    5389933,
-    5389937,
-    5389941,
-    5389944
+    5389973,
+    5389976,
+    5389979,
+    5389983,
+    5389987
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 295,
+  "inner_step": 298,
   "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:ce9d70c294e249b4ccf2db41a22f38d50f8b34d87dd84dea64d5a27b80eb84d3
+oid sha256:d030d64426e24d5f0fbded08bbec7825057932dfca81def57666ab506d6ac185
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3bd7251fd27c5d534e6fc4a2087d30a3fa0471ccca8a2b2dc3116a7e74facdf2
+oid sha256:7b33ab1f9fbb97acef685959e5e307b29ac14c65359f41878e246b47cc2861e2
 size 4040701744