chihoonlee10/T3Q-ko-solar-dpo-v1.0

Hugging Face
TEXT GENERATIONConcurrency Cost:1Model Size:10.7BQuant:FP8Ctx Length:4kLicense:apache-2.0Architecture:Transformer Open Weights Warm

T3Q-ko-solar-dpo-v1.0 is a 10.7 billion parameter language model developed by Chihoon Lee and T3Q. It is a DPO fine-tuned version of the davidkim205/nox-solar-10.7b-v4 model, featuring a 4096 token context length. This model is optimized for enhanced performance through Direct Preference Optimization.

Loading preview...

Overview

T3Q-ko-solar-dpo-v1.0 is a 10.7 billion parameter language model developed by Chihoon Lee and T3Q. This model is built upon the davidkim205/nox-solar-10.7b-v4 architecture and has undergone further fine-tuning using Direct Preference Optimization (DPO). It is designed to leverage the strengths of its base model while incorporating preference-based learning to refine its outputs.

Key Capabilities

  • DPO Fine-tuning: Utilizes Direct Preference Optimization to align model behavior with desired outcomes, potentially leading to more coherent and preferred responses.
  • Base Model Enhancement: Builds upon the davidkim205/nox-solar-10.7b-v4, suggesting a foundation in general language understanding and generation.
  • Context Length: Supports a context window of 4096 tokens, allowing for processing and generating moderately long sequences of text.

Good For

  • Applications requiring a model with improved alignment through DPO.
  • Tasks where the base davidkim205/nox-solar-10.7b-v4 model is suitable, with an expectation of refined performance.
  • Research and development in DPO techniques on existing large language models.

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