Run 6. Outer Step 20. Inner Step 20.

config.json

@@ -264,16 +264,15 @@
   "attention_bias": false,
   "attention_dropout": 0.0,
   "block_list": [
-    6138933,
-    6138947,
-    6138960
+    6138989,
+    6139004
   ],
   "bos_token_id": 1,
   "eos_token_id": 2,
   "hidden_act": "silu",
   "hidden_size": 2048,
   "initializer_range": 0.02,
-  "inner_step": 13,
+  "inner_step": 20,
   "intermediate_size": 5632,
   "last_allreduce_block": 6138837,
   "max_position_embeddings": 2048,

gradients.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c710a4e4b6080bdd0feccb0faa975b2950a1b5d210993d7a25eb833b68ac1171
+oid sha256:2f93d2e16af46c92e93d60278e7e098d8d71bf6bbc422f18e2c00a1e9aac0af2
 size 88389626

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:fbdec9153e0424f3cadeafcd2006eb66a96706d85fbf30afd379df9458910b88
+oid sha256:25703c8c60e2f76d51f9366cf44925e79659e22055083774a3e8474e294d8d8a
 size 8800564076

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c80cf531524a5a27d4943a781a9c9a4969fa545dd3d527b5dfac43e5df056b3d
+oid sha256:a75d72e0f7ef965d62389e84dd5998ce6b4e61f3f6a508b7f9b148f0734be8bc
 size 4400216536