Run 4. Outer Step 5. Inner Step 4.

config.json

@@ -268,22 +268,18 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5487947,
-    5487949,
-    5487952,
-    5487954,
-    5487956,
-    5487957,
-    5487959,
-    5487962,
-    5487964
+    5487966,
+    5487968,
+    5487971,
+    5487974,
+    5487977
   ],
   "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": 5487765,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:391f0276a2fed1268ae26b1901eb50f0c5c8429f8a1858b164892040cc5b7a18
+oid sha256:63097410adb97415c7397a1a2d4f1c7d733217857ce92a3e4ff327d526f8d3d4
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:95745ebf679183be951d7587d743032f65c44cbaecb987c9709272adc35f35a9
+oid sha256:d525de4e75ffcbddfbb67c26ead2c88c7804a0dd54f7832d0180ecd9c2c3e8c7
 size 4040701744