File size: 1,648 Bytes
9631acd f71e2ce 9631acd f71e2ce 69e09e0 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 |
---
license: apache-2.0
datasets:
- datajuicer/redpajama-wiki-refined-by-data-juicer
- datajuicer/redpajama-arxiv-refined-by-data-juicer
- datajuicer/redpajama-c4-refined-by-data-juicer
- datajuicer/redpajama-book-refined-by-data-juicer
- datajuicer/redpajama-cc-2019-30-refined-by-data-juicer
- datajuicer/redpajama-cc-2020-05-refined-by-data-juicer
- datajuicer/redpajama-cc-2021-04-refined-by-data-juicer
- datajuicer/redpajama-cc-2022-05-refined-by-data-juicer
- datajuicer/redpajama-cc-2023-06-refined-by-data-juicer
- datajuicer/redpajama-pile-stackexchange-refined-by-data-juicer
- datajuicer/redpajama-stack-code-refined-by-data-juicer
- datajuicer/the-pile-nih-refined-by-data-juicer
- datajuicer/the-pile-europarl-refined-by-data-juicer
- datajuicer/the-pile-philpaper-refined-by-data-juicer
- datajuicer/the-pile-pubmed-abstracts-refined-by-data-juicer
- datajuicer/the-pile-pubmed-central-refined-by-data-juicer
- datajuicer/the-pile-freelaw-refined-by-data-juicer
- datajuicer/the-pile-hackernews-refined-by-data-juicer
---
This is a reference LLM from [Data-Juicer](https://github.com/alibaba/data-juicer).
The model architecture is LLaMA-1.3B and we adopt the [OpenLLaMA](https://github.com/openlm-research/open_llama) implementation.
The model is pre-trained on 50B tokens of Data-Juicer's refined RedPajama and Pile.
It achieves an average score of 31.21 over 16 HELM tasks, beating LLMs trained on original RedPajama and Pile datasets.
For more details, please refer to our [paper](https://arxiv.org/abs/2309.02033).
![exp_llama](https://img.alicdn.com/imgextra/i2/O1CN019WtUPP1uhebnDlPR8_!!6000000006069-2-tps-2530-1005.png) |