search

msakarvadia / llm_bias

Investigating if we can find circuits in LLMs that reinforce human-biases found in training data
MIT License
0 stars 0 forks source link

Restructure Model Prompts to elicit shorter answer #6

Open msakarvadia opened 2 months ago

msakarvadia commented 2 months ago

Need model to output answer in less than 100 tokens.