liuda1/Mistral-7B-golden

TEXT GENERATIONConcurrency Cost:1Model Size:7BQuant:FP8Ctx Length:8kPublished:Dec 12, 2023License:unknownArchitecture:Transformer Cold

liuda1/Mistral-7B-golden is a 7 billion parameter language model based on the Mistral-7B-v0.1 architecture. It has been fine-tuned with additional English chat datasets and further reinforced with specific datasets. This model demonstrates enhanced chat capabilities, particularly in English, and is intended for general conversational AI applications.

Loading preview...

Model Overview

liuda1/Mistral-7B-golden is a 7-billion parameter language model built upon the Mistral-7B-v0.1 base architecture. It has undergone a multi-stage training process to enhance its conversational abilities.

Key Capabilities

  • Enhanced English Chat: The model has been fine-tuned with additional English chat datasets, resulting in improved performance in English-language conversations.
  • Reinforcement Training: Further reinforcement training on specific datasets has contributed to its overall chat proficiency.
  • Foundational Mistral-7B-v0.1: Leverages the robust architecture of Mistral-7B-v0.1 as its base.

Future Development

The developers plan to continue training this model to further improve its Chinese chat capabilities, indicating an ongoing commitment to multilingual conversational AI.

Good For

  • General English conversational AI tasks.
  • Applications requiring a 7B parameter model with refined chat abilities.

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