tmnam20/Llama-3.2-1B-Instruct-DPO

README.md

@@ -35,8 +35,8 @@ This model was trained with DPO, a method introduced in [Direct Preference Optim
 ### Framework versions
 
 - TRL: 0.15.2
-- Transformers: 4.49.0
-- Pytorch: 2.5.1+cu121
+- Transformers: 4.51.3
+- Pytorch: 2.6.0
 - Datasets: 3.3.2
 - Tokenizers: 0.21.0
 

adapter_config.json

@@ -23,13 +23,13 @@
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
+    "gate_proj",
+    "down_proj",
     "v_proj",
     "o_proj",
-    "k_proj",
-    "gate_proj",
     "q_proj",
     "up_proj",
-    "down_proj"
+    "k_proj"
   ],
   "task_type": "CAUSAL_LM",
   "use_dora": false,

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:aaf31c1e20eb1b17bca54945abbf56d22a4676a51d891ae86a0bf53abb6f9d29
+oid sha256:905de620b71252d5ddd00b2089a5a672551e90f14b9c418fe5e7ba7fcb33215e
 size 45118424

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2df0affef6309799c1ec5b85c44f3f04e54fbb8460f617e88914875ddb05349c
-size 6264
+oid sha256:afdf7e4b312599b9821bbf632aed05d8d42cb68cb7bba0504532e994b83fa5a2
+size 6200