zhangguohua/grpo_saved_lora

TEXT GENERATIONConcurrent Unit Cost:1Model Size:8BQuant:FP8Context Size:32kTool Calling:SupportedPublished:Mar 1, 2025License:apache-2.0Architecture:Transformer Open Weights Featherless Exclusive Cold

The zhangguohua/grpo_saved_lora is an 8 billion parameter Llama 3.1 instruction-tuned model, developed by zhangguohua and fine-tuned from unsloth/meta-llama-3.1-8b-instruct-unsloth-bnb-4bit. This model was trained using Unsloth and Huggingface's TRL library, enabling 2x faster training. It is optimized for efficient performance and is suitable for applications requiring a capable Llama 3.1 base with enhanced training speed.

Loading preview...

Model Overview

The zhangguohua/grpo_saved_lora is an 8 billion parameter Llama 3.1 instruction-tuned model, developed by zhangguohua. It is fine-tuned from the unsloth/meta-llama-3.1-8b-instruct-unsloth-bnb-4bit base model, leveraging the Unsloth library for accelerated training.

Key Characteristics

  • Base Model: Fine-tuned from Meta Llama 3.1 8B Instruct.
  • Training Efficiency: Utilizes Unsloth and Huggingface's TRL library, resulting in 2x faster training compared to standard methods.
  • Developer: zhangguohua.
  • License: Released under the Apache-2.0 license.

Use Cases

This model is particularly well-suited for:

  • Applications requiring a Llama 3.1 8B instruction-tuned model.
  • Scenarios where efficient and faster fine-tuning is a priority.
  • Projects benefiting from the performance characteristics of the Llama 3.1 architecture.