Downloads · 30 days
10
14% of all-time downloads
safal312/BI-Gemma-7layers-SGR-Dolci
BI-Gemma-7layers-SGR-Dolci is a text generation model from safal312. Use it when you need the model to write or continue text. The card lists the license as apache-2.0.
This model is a depth-pruned version of Gemma-2B-IT, obtained via Block Influence (BI) layer pruning and post-trained using QLoRA with Self-Generated Responses (SGR) on the Dolci dataset.
Downloads · 30 days
10
14% of all-time downloads
All-time downloads
70
Public
Parameters
2.1B
4.2 GB on disk
Likes
0
Public
Click a slice to open those files.
.safetensors4.1 GB · 99%
From the Hugging Face model README
This model is a depth-pruned version of Gemma-2B-IT, obtained via Block Influence (BI) layer pruning and post-trained using QLoRA with Self-Generated Responses (SGR) on the Dolci dataset.
It was released as part of our study on the limits of layer pruning for generative reasoning.
This checkpoint is intended for research and analysis of pruning and recovery, not as a production model.
On the Limits of Layer Pruning for Generative Reasoning in LLMs
https://arxiv.org/abs/2602.01997
@misc{shrestha2026limitslayerpruninggenerative,
title={On the Limits of Layer Pruning for Generative Reasoning in LLMs},
author={Safal Shrestha and Anubhav Shrestha and Aadim Nepal and Minwu Kim and Keith Ross},
year={2026},
eprint={2602.01997},
archivePrefix={arXiv},
primaryClass={cs.LG},
url={https://arxiv.org/abs/2602.01997},
}