Refusal Token Models
Collection
This collection contains models described in the refusal token paper published in COLM 2025.
•
4 items
•
Updated
Model with Multiple Refusal Tokens For Each Category
This model is a fine-tuned version of meta-llama/Meta-Llama-3-8B on UltraChat SFT with the respond token, CoCoNoT refusals with the refuse token, and CoCoNoT's contrast data as SFT data with the multiple refusal tokens for each of the five categories. Note that this model is not the model found in the paper the original models are not able to be released due to corporate legalities.
Suggusted Method For Generations
For generating a output from this model, please refer to the code found in repo in the coconot_eval folder.
- Downloads last month
- 64
Inference Providers
NEW
This model isn't deployed by any Inference Provider.
🙋
Ask for provider support