Downloads · 30 days
9
18% of all-time downloads
D1zzYzz/GRIT-GSM8K-QLORA-llama-3.1-8B-Energy-0.9
GRIT-GSM8K-QLORA-llama-3.1-8B-Energy-0.9 is a text generation model from D1zzYzz. Use it when you need the model to write or continue text. It is set up for peft. The card lists the license as apache-2.0.
This model is a fine-tuned version of meta-llama/Llama-3.1-8B using the GRIT (Geometric Reprojection Instruction Tuning) algorithm and QLoRA on the openai/gsm8k dataset.
Downloads · 30 days
9
18% of all-time downloads
All-time downloads
50
Public
Repo size
353 MB
Likes
0
Public
Click a slice to open those files.
.safetensors336 MB · 95%
From the Hugging Face model README
This model is a fine-tuned version of meta-llama/Llama-3.1-8B using the GRIT (Geometric Reprojection Instruction Tuning) algorithm and QLoRA on the openai/gsm8k dataset.
The base model is quantized to 4-bit (NF4) and optimized with Unsloth to enable efficient fine-tuning.
In benchmark comparisons, GRIT has shown faster convergence and better stability than standard LoRA or fine-tuning, making it well-suited for efficient single-epoch training. The use of Unsloth further accelerates this process.
If you use this model, please cite the original GRIT paper and:
@misc{grit-lora-Llama-3.1-8B-gsm8k},
title={ meta-llama/Llama-3.1-8B Fine-tuned with GRIT on openai/gsm8k },
author={D1zzYzz},
year={2025},
publisher={Hugging Face},
url={https://huggingface.co/D1zzYzz/GRIT-GSM8K-QLORA-llama-3.1-8B-Energy-0.9}
}
This model inherits the Apache 2.0 license.