NeuralNovel/Ignis-7B-DPO

TEXT GENERATIONConcurrency Cost:1Model Size:7BQuant:FP8Ctx Length:8kPublished:Feb 24, 2024License:apache-2.0Architecture:Transformer0.0K Open Weights Cold

Ignis-7B-DPO is a 7 billion parameter language model developed by NeuralNovel, fine-tuned using Direct Preference Optimization (DPO) on the Neural-DPO dataset. This model is designed for general language generation tasks, leveraging its DPO training for improved response quality and alignment. With an 8192-token context length, it offers robust performance for various applications requiring nuanced text understanding and generation.

Loading preview...

Ignis-7B-DPO: A DPO-Tuned 7B Language Model

Ignis-7B-DPO is a 7 billion parameter language model developed by NeuralNovel, distinguished by its training methodology. The model was fine-tuned using Direct Preference Optimization (DPO) on the proprietary Neural-DPO dataset, utilizing A-100 80GB GPUs for this process. This DPO approach aims to enhance the model's ability to generate high-quality, aligned, and preferred responses based on human feedback.

Key Capabilities

  • Direct Preference Optimization (DPO): Leverages DPO for improved response quality and alignment.
  • General Language Generation: Suitable for a wide array of text generation tasks.
  • 7 Billion Parameters: Offers a balance of performance and computational efficiency.
  • 8192-Token Context Length: Provides ample context for complex queries and longer interactions.

Good For

  • Applications requiring a model with enhanced alignment and preference-based tuning.
  • General-purpose text generation and understanding tasks.
  • Developers seeking a 7B model trained with advanced fine-tuning techniques.

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