jondurbin/airoboros-13b
jondurbin/airoboros-13b is a 13 billion parameter LLaMA-based model fine-tuned by jondurbin using a completely synthetic dataset. This model was an experimental attempt to generate a broader range of data, including potentially 'harmful' content, by employing a 'jailbreak' prompt with OpenAI's GPT-4 and GPT-3.5-turbo. It achieves a GPT-3.5 adjusted score of 98.087 in internal evaluations, indicating strong performance relative to GPT-3.5, but the developer advises against its general use due to output quality and potential harmful content.
Hugging Face Popular Sampler Settings
Most commonly used values from Featherless users
temperature
This setting influences the sampling randomness. Lower values make the model more deterministic; higher values introduce randomness. Zero is greedy sampling.
–
top_p
This setting controls the cumulative probability of considered top tokens. Must be in (0, 1]. Set to 1 to consider all tokens.
–
top_k
This limits the number of top tokens to consider. Set to -1 to consider all tokens.
–
frequency_penalty
This setting penalizes new tokens based on their frequency in the generated text. Values > 0 encourage new tokens; < 0 encourages repetition.
–
presence_penalty
This setting penalizes new tokens based on their presence in the generated text so far. Values > 0 encourage new tokens; < 0 encourages repetition.
–
repetition_penalty
This setting penalizes new tokens based on their appearance in the prompt and generated text. Values > 1 encourage new tokens; < 1 encourages repetition.
–
min_p
This setting representing the minimum probability for a token to be considered relative to the most likely token. Must be in [0, 1]. Set to 0 to disable.
–