DrNicefellow/ChatAllInOne-Yi-34B-200K-V1

TEXT GENERATIONConcurrency Cost:2Model Size:34BQuant:FP8Ctx Length:32kPublished:Jan 30, 2024License:apache-2.0Architecture:Transformer0.0K Open Weights Cold

DrNicefellow/ChatAllInOne-Yi-34B-200K-V1 is a 34 billion parameter chat language model, fine-tuned from the 01-ai/Yi-34B-200K base model using QLoRA and the unsloth tool. Optimized on the CHAT-ALL-IN-ONE-v1 dataset, it excels in diverse conversational scenarios, maintaining context and coherence over extended dialogues. This model is designed for enhanced understanding and generation of conversational language across various chat applications.

Loading preview...

Overview

DrNicefellow/ChatAllInOne-Yi-34B-200K-V1 is a 34 billion parameter chat language model, built upon the 01-ai/Yi-34B-200K base model. It was fine-tuned using the QLoRA technique with the unsloth tool, leveraging the specialized CHAT-ALL-IN-ONE-v1 dataset. This fine-tuning process aimed to optimize the model specifically for conversational tasks.

Key Capabilities

  • Enhanced Conversational Understanding: Improved ability to interpret and respond to diverse conversational nuances.
  • Coherent Dialogue Generation: Maintains context and logical flow over longer chat interactions.
  • Versatile Chat Performance: Designed to perform well across various chat scenarios, including casual, formal, and domain-specific discussions.

Use Cases

This model is particularly well-suited for applications requiring robust and coherent conversational AI. Its fine-tuning on a comprehensive chat dataset makes it a strong candidate for chatbots, virtual assistants, and interactive dialogue systems where maintaining context and generating natural language responses are critical. The model utilizes the Vicuna 1.1 prompt format for interaction.

Finetunes
5 models