uukuguy/speechless-mistral-hermes-code-7b

TEXT GENERATIONConcurrency Cost:1Model Size:7BQuant:FP8Ctx Length:8kPublished:Feb 6, 2024License:apache-2.0Architecture:Transformer0.0K Open Weights Cold

The uukuguy/speechless-mistral-hermes-code-7b is a 7 billion parameter language model based on the Mistral-7B-v0.1 architecture, fine-tuned by uukuguy. It is specifically optimized for improving reasoning and planning abilities, particularly in code-related tasks. This model leverages a diverse dataset including OpenHermes-2.5, Python evaluation instructions, Spider, and Evol-instruction-66k, making it suitable for programming assistance and code generation.

Loading preview...

Model Overview

The uukuguy/speechless-mistral-hermes-code-7b is a 7 billion parameter language model developed by uukuguy, built upon the mistralai/Mistral-7B-v0.1 base. Its primary objective is to enhance the model's reasoning and planning capabilities, especially within programming contexts.

Key Capabilities & Training

This model was fine-tuned using a comprehensive dataset totaling 986,000 samples, which includes:

  • teknium/OpenHermes-2.5: A high-quality instruction-following dataset.
  • TokenBender/python_eval_instruct_51k: Focused on Python evaluation and instruction.
  • Spider: A dataset for text-to-SQL tasks, improving logical reasoning.
  • codefuse-ai/Evol-instruction-66k: Likely contributes to instruction-following and code generation.

It supports an 8192-token context length and is designed to accept the Alpaca instruction format, making it straightforward to prompt for various tasks.

Use Cases

  • Programming Assistance: Excels at understanding and generating code, particularly for tasks requiring logical reasoning and planning.
  • Code Implementation: Capable of implementing code snippets or full functions based on instructions.
  • Instruction Following: Designed to respond effectively to detailed instructions, especially in technical domains.

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