Avelina/lovelace-medium-alpha1
Text Generation
•
Updated
•
17
•
1
Direct Preference Heads (Preprint)
This collection contains the pre-trained, fine-tuned and aligned models for the Direct Preference Heads paper.
Note Pretrained Transformer-XL model with 550M parameters, trained on 100B tokens from The Pile.
-
Avelina/lovelace-medium-alpha1-sft
Text Generation • Updated • 12 -
Avelina/lovelace-medium-alpha1-dph
Updated • 11 • 1 -
Would I Lie To You? Inference Time Alignment of Language Models using Direct Preference Heads
Paper • 2405.20053 • Published • 2