DANWPDO/dpo_project

README.md

@@ -35,14 +35,14 @@ More information needed
 ### Training hyperparameters
 
 The following hyperparameters were used during training:
-- learning_rate: 1
+- learning_rate: 0.0002
 - train_batch_size: 1
 - eval_batch_size: 8
 - seed: 42
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
 - lr_scheduler_warmup_steps: 2
-- training_steps: 1
+- training_steps: 10
 
 ### Framework versions
 

adapter_config.json

@@ -20,8 +20,8 @@
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "v_pro",
-    "q_proj"
+    "q_proj",
+    "v_pro"
   ],
   "task_type": "CAUSAL_LM",
   "use_dora": false,

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:727553dfd76e9fa6a9fe99867590b8fca9e8730156d39bf9c1966edf565246b1
+oid sha256:a836f4658e0a44814307df2844f2dc3763d1497d3952ccd58c84fca058101627
 size 8397056

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:133a226e1b0a80338b25148faf2269ba1ace18152e5b3c73c7a0070baf9a05fc
+oid sha256:e5ec58ef909997203074211940053e20eea1462c0b6d1048b7e1f0b2847e5aaf
 size 4411