File size: 1,448 Bytes
aa7b106 |
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 33 34 35 36 37 38 |
Quantization made by Richard Erkhov.
[Github](https://github.com/RichardErkhov)
[Discord](https://discord.gg/pvy7H8DZMG)
[Request more models](https://github.com/RichardErkhov/quant_request)
fairseq-dense-2.7B - bnb 8bits
- Model creator: https://huggingface.co/KoboldAI/
- Original model: https://huggingface.co/KoboldAI/fairseq-dense-2.7B/
Original model description:
---
language: en
---
This is a Hugging Face transformers-compatible conversion of the original dense 2.7B-parameter model from the paper "[Efficient Large Scale Language Modeling with Mixtures of Experts](https://arxiv.org/abs/2112.10684)" from Artetxe et al. Please refer to the original model card, which can be found at https://github.com/facebookresearch/fairseq/blob/main/examples/moe_lm/model_card.md.
# [Open LLM Leaderboard Evaluation Results](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard)
Detailed results can be found [here](https://huggingface.co/datasets/open-llm-leaderboard/details_KoboldAI__fairseq-dense-2.7B)
| Metric | Value |
|-----------------------|---------------------------|
| Avg. | 33.67 |
| ARC (25-shot) | 33.79 |
| HellaSwag (10-shot) | 65.74 |
| MMLU (5-shot) | 26.44 |
| TruthfulQA (0-shot) | 34.57 |
| Winogrande (5-shot) | 63.93 |
| GSM8K (5-shot) | 0.0 |
| DROP (3-shot) | 11.24 |
|