|
--- |
|
license: apache-2.0 |
|
datasets: |
|
- gair-prox/RedPajama-pro |
|
language: |
|
- en |
|
tags: |
|
- math |
|
- reasoning |
|
--- |
|
|
|
# ProX-RedPJ-0.3B |
|
|
|
<p align="center"> |
|
<img src="prox.png" width="200"> |
|
</p> |
|
|
|
[ArXiv](http://arxiv.org/abs/xxxx) | [Models](https://huggingface.co/gair-prox/ProX-RedPJ-0.3B) | [Data](https://huggingface.co/datasets/gair-prox/RedPajama-pro) | [Code](https://github.com/GAIR-NLP/program-every-example) |
|
|
|
**ProX-RedPJ-0.3B** is a tiny language model. It was and trained on the [RedPajama-V2-pro](https://huggingface.co/datasets/gair-prox/RedPajama-pro) for 25B tokens. |
|
|
|
## Evaluations |
|
|
|
ProX models are evaluated over 10 language model benchmarks in zero-shot setting. |
|
|
|
| | ARC-C | ARC-E | CSQA | HellaS | MMLU | OBQA | PiQA | SIQA | WinoG | SciQ | AVG | |
|
|-----------------------|-------|-------|-------|-----------|-------|-------|-------|-------|-------|-------|------| |
|
| raw | 22.6% | 41.9% | 29.7% | 32.8% | 26.2% | 26.4% | 62.2% | 39.3% | 51.3% | 63.3% | 39.6 | |
|
| ours | 25.9% | 47.5% | 29.2% | 36.7% | 28.1% | 30.2% | 64.6% | 38.0% | 51.7% | 71.4% | 42.3 | |
|
|
|
### Citation |
|
``` |
|
@misc{TBD |
|
} |
|
``` |
