test
This commit is contained in:
@@ -8,6 +8,7 @@ lora_alpha: 32
|
|||||||
lora_dropout: 0.1
|
lora_dropout: 0.1
|
||||||
lora_target_linear: true
|
lora_target_linear: true
|
||||||
rl: dpo
|
rl: dpo
|
||||||
|
dpo_use_weighting: true
|
||||||
|
|
||||||
datasets:
|
datasets:
|
||||||
- path: arcee-ai/distilabel-intel-orca-dpo-pairs-binarized
|
- path: arcee-ai/distilabel-intel-orca-dpo-pairs-binarized
|
||||||
|
|||||||
Reference in New Issue
Block a user