mahiatlinux/MasherAI-v6.1-7B-checkpoint3-code3

TEXT GENERATIONConcurrency Cost:1Model Size:7BQuant:FP8Ctx Length:8kPublished:Apr 5, 2024License:apache-2.0Architecture:Transformer Open Weights Cold

The mahiatlinux/MasherAI-v6.1-7B-checkpoint3-code3 is a 7 billion parameter Mistral-based language model, fine-tuned from mahiatlinux/MasherAI-v6.1-7B-checkpoint3-code2. This model was trained using Unsloth and Huggingface's TRL library, achieving 2x faster training speeds. With an 8192-token context length, it is optimized for tasks benefiting from efficient fine-tuning and a Mistral architecture.

Loading preview...

Model Overview

mahiatlinux/MasherAI-v6.1-7B-checkpoint3-code3 is a 7 billion parameter language model developed by mahiatlinux. It is a Mistral-based model, specifically fine-tuned from mahiatlinux/MasherAI-v6.1-7B-checkpoint3-code2. A key characteristic of this model's development is its training methodology, which leveraged Unsloth and Huggingface's TRL library to achieve a reported 2x faster training speed.

Key Capabilities

  • Efficient Training: Benefits from Unsloth's optimizations for faster fine-tuning.
  • Mistral Architecture: Inherits the robust capabilities of the Mistral model family.
  • Context Length: Supports an 8192-token context window, suitable for tasks requiring moderate input lengths.

Good For

  • Developers looking for a Mistral-based model that has undergone efficient fine-tuning.
  • Applications where faster training iteration cycles are beneficial.
  • Tasks that align with the general capabilities of 7B parameter Mistral models, potentially with a focus on code-related applications given its checkpoint name.

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