NCSOFT/Llama-3-OffsetBias-8B
TEXT GENERATIONConcurrency Cost:1Model Size:8BQuant:FP8Ctx Length:8kPublished:Jul 11, 2024License:llama3Architecture:Transformer0.0K Cold

NCSOFT/Llama-3-OffsetBias-8B is an 8 billion parameter generative judge model developed by NC Research, built upon Meta Llama-3-8B-Instruct. It is specifically fine-tuned for pairwise preference evaluation tasks, designed to be robust against common evaluation biases. This model excels at identifying the better of two outputs for a given instruction, making it ideal for debiased AI model evaluation.

Loading preview...

Popular Sampler Settings

Top 3 parameter combinations used by Featherless users for this model. Click a tab to see each config.

temperature
top_p
top_k
frequency_penalty
presence_penalty
repetition_penalty
min_p