Run 4. Outer Step 2. Inner Step 3.

config.json

@@ -268,20 +268,20 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5480865,
-    5480869,
-    5480873,
-    5480877,
-    5480881,
-    5480885,
-    5480889
+    5480893,
+    5480896,
+    5480901,
+    5480905,
+    5480909,
+    5480912,
+    5480916
   ],
   "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": 5480541,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:eb8cb2d49681b8677a1ff6398e69b8a8079b294c6fef163371aecfe358fcd054
+oid sha256:89c702988b00529307d376d124e7283f6657aeb6e6e75fa3f95fbc6b66c4edf6
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:cc1145a7c7c8b07fe2e416728e2ca92c40c8104585fef90f4a5594265b20baff
+oid sha256:b5398afe5ad7a603c127d58e81d41dea31e30dc51c62ccfd94b9d4369b906a1c
 size 4040701744