Base Models Beat Aligned Models at Randomness and Creativity
Paper • 2505.00047 • Published • 3
StageHand-12B
Experimental RP model - Now more "steerable"
- This is an experimental model that can only do RP. A continuation from TinyRP-12B; but more steerable than it.
- Use ChatML as format in "Text completion" mode on ST.
- Use 'default' hyperparameters. (Temp= 0.7, etc. personally I also use rep pen, but DRY may work better for you)
This model is a bit mroe capable of being used as a true "assistant" or "agent".
Feedback very welcome!
Why train on a base model?
According to this paper: Base Models Beat Aligned Models at Randomness and Creativity; and to avoid any possible "GPT-isms", I decided to train on a base model. Think of it as more mallable clay vs re-shaping something that was already formed to be something else.
This is what led to the behavior observed in this model, where the model just legitimately doesn't understand being an "assistant" outside of being a character that is an assistant. SO while the model is probably not useful outside of RP, it is also not intended to be.
- Downloads last month
- 3
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support
Model tree for DarwinAnim8or/Stagehand-12-old
Base model
mistralai/Mistral-Nemo-Base-2407