kaitchup/Mayonnaise-4in1-022

TEXT GENERATIONConcurrency Cost:1Model Size:7BQuant:FP8Ctx Length:8kPublished:Jan 27, 2024License:apache-2.0Architecture:Transformer0.0K Open Weights Cold

kaitchup/Mayonnaise-4in1-022 is a 7 billion parameter causal language model developed by The Kaitchup, based on mistralai/Mistral-7B-v0.1. This model is a mixture of experts created using mergekit, combining several Mistral-7B variants. It is optimized for general language tasks and has achieved a top ranking on the Open LLM Leaderboard among 7B models, making it suitable for a wide range of English NLP applications.

Loading preview...

Model Overview

kaitchup/Mayonnaise-4in1-022 is a 7 billion parameter causal language model, developed by The Kaitchup. It is constructed as a "mixture of experts" using mergekit, building upon the mistralai/Mistral-7B-v0.1 architecture. This model integrates parameters from mncai/mistral-7b-dpo-v5, FelixChao/WestSeverus-7B-DPO-v2, and BarryFutureman/NeuralTurdusVariant1-7B to enhance its capabilities.

Key Characteristics

  • Architecture: Causal language model based on Mistral-7B-v0.1.
  • Development Method: Created via a TIES-method merge using mergekit, combining multiple fine-tuned Mistral-7B variants.
  • Performance: Ranked first on the Open LLM Leaderboard among 7B models as of January 28th, 2024, though the developer notes potential evaluation benchmark contamination from merged models.
  • License: Released under the Apache 2.0 License.

Potential Use Cases

This model is suitable for a variety of general English natural language processing tasks, including text generation, summarization, and question answering, leveraging its strong performance on public leaderboards. Developers should be aware of the potential for benchmark contamination when interpreting performance metrics.

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