mgoin's picture
Update README.md
e2f62cf verified
---
license: other
license_name: nvidia-open-model-license
license_link: >-
https://developer.download.nvidia.com/licenses/nvidia-open-model-license-agreement-june-2024.pdf
tags:
- fp8
- vllm
base_model: nvidia/Minitron-4B-Base
---
# Minitron-4B-Base-FP8
FP8 quantized checkpoint of [nvidia/Minitron-4B-Base](https://huggingface.co/nvidia/Minitron-4B-Base) for use with vLLM.
```
lm_eval --model vllm --model_args pretrained=mgoin/Minitron-4B-Base-FP8 --tasks gsm8k --num_fewshot 5 --batch_size auto
vllm (pretrained=mgoin/Minitron-4B-Base-FP8), gen_kwargs: (None), limit: None, num_fewshot: 5, batch_size: auto
|Tasks|Version| Filter |n-shot| Metric | |Value | |Stderr|
|-----|------:|----------------|-----:|-----------|---|-----:|---|-----:|
|gsm8k| 3|flexible-extract| 5|exact_match|↑ |0.2305|± |0.0116|
| | |strict-match | 5|exact_match|↑ |0.2282|± |0.0116|
```