Run 4. Outer Step 33. Inner Step 4.

config.json

@@ -268,18 +268,18 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5546782,
-    5546786,
-    5546790,
-    5546794,
-    5546798
+    5546803,
+    5546807,
+    5546811,
+    5546816,
+    5546820
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 3,
+  "inner_step": 4,
   "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 sha256:050e63b99f1153537ae6806efbec3ff4c8c99f5c189260ba7c8eb21138765a9f
+oid sha256:76543a67761f15f70fb5a74b1541cff45ede3cccdf76dd87c189e38a5d0d4f55
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:105a4265e62473701bf7ad6f865a49597828ff56847b0afa003f4c6290b75ec1
+oid sha256:6eebb2511f3302082650e04d44ca76c2cc71146f5822b5918b2e045f546dc650
 size 4040701744