You can deploy this model and start earning money today!

RichardErkhov/
MiniLLM_-_MiniPLM-Qwen-500M-exl2

Name: MiniLLM_-_MiniPLM-Qwen-500M-exl2
Availability: InStock

2410.17215

Views

No views yet

Quantization made by Richard Erkhov.

Github

Discord

Request more models

MiniPLM-Qwen-500M - EXL2

Model creator: https://huggingface.co/MiniLLM/
Original model: https://huggingface.co/MiniLLM/MiniPLM-Qwen-500M/

Available sizes

Branch	Bits	Description
8_0	8.0	Maximum quality that ExLlamaV2 can produce, near unquantized performance.
6_5	6.5	Very similar to 8.0, good tradeoff of size vs performance, recommended.
5_0	5.0	Slightly lower quality vs 6.5, but usable
4_25	4.25	GPTQ equivalent bits per weight, slightly higher quality.
3_5	3.5	Lower quality, only use if you have to.

Download instructions

With git:

git clone --single-branch --branch 6_5 https://huggingface.co/MiniLLM_-_MiniPLM-Qwen-500M-exl2 MiniPLM-Qwen-500M-6_5

With huggingface hub:

pip3 install huggingface-hub

To download a specific branch, use the --revision parameter. For example, to download the 6.5 bpw branch: Linux:

huggingface-cli download MiniLLM_-_MiniPLM-Qwen-500M-exl2 --revision 6_5 --local-dir MiniPLM-Qwen-500M-6_5 --local-dir-use-symlinks False

Windows (which apparently doesn't like _ in folders sometimes?):

huggingface-cli download MiniLLM_-_MiniPLM-Qwen-500M-exl2 --revision 6_5 --local-dir MiniPLM-Qwen-500M-6.5 --local-dir-use-symlinks False

Original model description:

library_name: transformers license: apache-2.0 datasets:

monology/pile-uncopyrighted
MiniLLM/pile-diff_samp-qwen_1.8B-qwen_104M-r0.5 language:
en metrics:
accuracy pipeline_tag: text-generation

MinPLM-Qwen-500M

paper | code

MiniPLM-Qwen-500M is a 500M model with Qwen achitecture pre-trained from scratch on the Pile using the MiniPLM knowledge distillation framework with the offcial Qwen1.5-1.8B as the teacher model.

We also open-source the pre-training corpus refined by Difference Sampling in MiniPLM for reproducibility.

Evaluation

MiniPLM models achieves better performance given the same computation and scales well across model sizes:

Baseline Models

Citation

bibtex

@article{miniplm,
    title={MiniPLM: Knowledge Distillation for Pre-Training Language Models}, 
    author={Yuxian Gu and Hao Zhou and Fandong Meng and Jie Zhou and Minlie Huang},
    journal={arXiv preprint arXiv:2410.17215},
    year={2024}
}

RichardErkhov/MiniLLM_-_MiniPLM-Qwen-500M-exl2