README.md

Quantization made by Richard Erkhov.

[Github](https://github.com/RichardErkhov)

[Discord](https://discord.gg/pvy7H8DZMG)

[Request more models](https://github.com/RichardErkhov/quant_request)


Taiwan-LLM-7B-v2.0-base - GGUF
- Model creator: https://huggingface.co/yentinglin/
- Original model: https://huggingface.co/yentinglin/Taiwan-LLM-7B-v2.0-base/


| Name | Quant method | Size |
| ---- | ---- | ---- |
| [Taiwan-LLM-7B-v2.0-base.Q2_K.gguf](https://huggingface.co/RichardErkhov/yentinglin_-_Taiwan-LLM-7B-v2.0-base-gguf/blob/main/Taiwan-LLM-7B-v2.0-base.Q2_K.gguf) | Q2_K | 2.36GB |
| [Taiwan-LLM-7B-v2.0-base.IQ3_XS.gguf](https://huggingface.co/RichardErkhov/yentinglin_-_Taiwan-LLM-7B-v2.0-base-gguf/blob/main/Taiwan-LLM-7B-v2.0-base.IQ3_XS.gguf) | IQ3_XS | 2.6GB |
| [Taiwan-LLM-7B-v2.0-base.IQ3_S.gguf](https://huggingface.co/RichardErkhov/yentinglin_-_Taiwan-LLM-7B-v2.0-base-gguf/blob/main/Taiwan-LLM-7B-v2.0-base.IQ3_S.gguf) | IQ3_S | 2.75GB |
| [Taiwan-LLM-7B-v2.0-base.Q3_K_S.gguf](https://huggingface.co/RichardErkhov/yentinglin_-_Taiwan-LLM-7B-v2.0-base-gguf/blob/main/Taiwan-LLM-7B-v2.0-base.Q3_K_S.gguf) | Q3_K_S | 2.75GB |
| [Taiwan-LLM-7B-v2.0-base.IQ3_M.gguf](https://huggingface.co/RichardErkhov/yentinglin_-_Taiwan-LLM-7B-v2.0-base-gguf/blob/main/Taiwan-LLM-7B-v2.0-base.IQ3_M.gguf) | IQ3_M | 2.9GB |
| [Taiwan-LLM-7B-v2.0-base.Q3_K.gguf](https://huggingface.co/RichardErkhov/yentinglin_-_Taiwan-LLM-7B-v2.0-base-gguf/blob/main/Taiwan-LLM-7B-v2.0-base.Q3_K.gguf) | Q3_K | 3.07GB |
| [Taiwan-LLM-7B-v2.0-base.Q3_K_M.gguf](https://huggingface.co/RichardErkhov/yentinglin_-_Taiwan-LLM-7B-v2.0-base-gguf/blob/main/Taiwan-LLM-7B-v2.0-base.Q3_K_M.gguf) | Q3_K_M | 3.07GB |
| [Taiwan-LLM-7B-v2.0-base.Q3_K_L.gguf](https://huggingface.co/RichardErkhov/yentinglin_-_Taiwan-LLM-7B-v2.0-base-gguf/blob/main/Taiwan-LLM-7B-v2.0-base.Q3_K_L.gguf) | Q3_K_L | 3.35GB |
| [Taiwan-LLM-7B-v2.0-base.IQ4_XS.gguf](https://huggingface.co/RichardErkhov/yentinglin_-_Taiwan-LLM-7B-v2.0-base-gguf/blob/main/Taiwan-LLM-7B-v2.0-base.IQ4_XS.gguf) | IQ4_XS | 3.4GB |
| [Taiwan-LLM-7B-v2.0-base.Q4_0.gguf](https://huggingface.co/RichardErkhov/yentinglin_-_Taiwan-LLM-7B-v2.0-base-gguf/blob/main/Taiwan-LLM-7B-v2.0-base.Q4_0.gguf) | Q4_0 | 3.56GB |
| [Taiwan-LLM-7B-v2.0-base.IQ4_NL.gguf](https://huggingface.co/RichardErkhov/yentinglin_-_Taiwan-LLM-7B-v2.0-base-gguf/blob/main/Taiwan-LLM-7B-v2.0-base.IQ4_NL.gguf) | IQ4_NL | 3.58GB |
| [Taiwan-LLM-7B-v2.0-base.Q4_K_S.gguf](https://huggingface.co/RichardErkhov/yentinglin_-_Taiwan-LLM-7B-v2.0-base-gguf/blob/main/Taiwan-LLM-7B-v2.0-base.Q4_K_S.gguf) | Q4_K_S | 3.59GB |
| [Taiwan-LLM-7B-v2.0-base.Q4_K.gguf](https://huggingface.co/RichardErkhov/yentinglin_-_Taiwan-LLM-7B-v2.0-base-gguf/blob/main/Taiwan-LLM-7B-v2.0-base.Q4_K.gguf) | Q4_K | 3.8GB |
| [Taiwan-LLM-7B-v2.0-base.Q4_K_M.gguf](https://huggingface.co/RichardErkhov/yentinglin_-_Taiwan-LLM-7B-v2.0-base-gguf/blob/main/Taiwan-LLM-7B-v2.0-base.Q4_K_M.gguf) | Q4_K_M | 3.8GB |
| [Taiwan-LLM-7B-v2.0-base.Q4_1.gguf](https://huggingface.co/RichardErkhov/yentinglin_-_Taiwan-LLM-7B-v2.0-base-gguf/blob/main/Taiwan-LLM-7B-v2.0-base.Q4_1.gguf) | Q4_1 | 3.95GB |
| [Taiwan-LLM-7B-v2.0-base.Q5_0.gguf](https://huggingface.co/RichardErkhov/yentinglin_-_Taiwan-LLM-7B-v2.0-base-gguf/blob/main/Taiwan-LLM-7B-v2.0-base.Q5_0.gguf) | Q5_0 | 4.33GB |
| [Taiwan-LLM-7B-v2.0-base.Q5_K_S.gguf](https://huggingface.co/RichardErkhov/yentinglin_-_Taiwan-LLM-7B-v2.0-base-gguf/blob/main/Taiwan-LLM-7B-v2.0-base.Q5_K_S.gguf) | Q5_K_S | 4.33GB |
| [Taiwan-LLM-7B-v2.0-base.Q5_K.gguf](https://huggingface.co/RichardErkhov/yentinglin_-_Taiwan-LLM-7B-v2.0-base-gguf/blob/main/Taiwan-LLM-7B-v2.0-base.Q5_K.gguf) | Q5_K | 4.45GB |
| [Taiwan-LLM-7B-v2.0-base.Q5_K_M.gguf](https://huggingface.co/RichardErkhov/yentinglin_-_Taiwan-LLM-7B-v2.0-base-gguf/blob/main/Taiwan-LLM-7B-v2.0-base.Q5_K_M.gguf) | Q5_K_M | 4.45GB |
| [Taiwan-LLM-7B-v2.0-base.Q5_1.gguf](https://huggingface.co/RichardErkhov/yentinglin_-_Taiwan-LLM-7B-v2.0-base-gguf/blob/main/Taiwan-LLM-7B-v2.0-base.Q5_1.gguf) | Q5_1 | 4.72GB |
| [Taiwan-LLM-7B-v2.0-base.Q6_K.gguf](https://huggingface.co/RichardErkhov/yentinglin_-_Taiwan-LLM-7B-v2.0-base-gguf/blob/main/Taiwan-LLM-7B-v2.0-base.Q6_K.gguf) | Q6_K | 5.15GB |
| [Taiwan-LLM-7B-v2.0-base.Q8_0.gguf](https://huggingface.co/RichardErkhov/yentinglin_-_Taiwan-LLM-7B-v2.0-base-gguf/blob/main/Taiwan-LLM-7B-v2.0-base.Q8_0.gguf) | Q8_0 | 6.67GB |




Original model description:
---
license: apache-2.0
language:
- zh
widget:
 - text: "A chat between a curious user and an artificial intelligence assistant. The assistant gives helpful, detailed, and polite answers to the user's questions. USER: 你好，請問你可以幫我寫一封推薦信嗎？ ASSISTANT:"
library_name: transformers
pipeline_tag: text-generation
extra_gated_heading: Acknowledge license to accept the repository.
extra_gated_prompt: Please contact the author for access.
extra_gated_button_content: Acknowledge license 同意以上內容
extra_gated_fields:
  Name: text
  Mail: text
  Organization: text
  Country: text
  Any utilization of the Taiwan LLM repository mandates the explicit acknowledgment and attribution to the original author: checkbox
  使用Taiwan LLM必須明確地承認和歸功於優必達株式會社 Ubitus 以及原始作者: checkbox
---
<img src="https://cdn-uploads.huggingface.co/production/uploads/5df9c78eda6d0311fd3d541f/CmusIT5OlSXvFrbTJ7l-C.png" alt="Taiwan LLM Logo" width="800" style="margin-left:'auto' margin-right:'auto' display:'block'"/>

# 🌟 Checkout [Taiwan-LLM Demo Chat-UI](http://www.twllm.com) 🌟

# Model Card for Taiwan LLM 7B v2.0 base

Taiwan LLM is an advanced language model tailored for Traditional Chinese, focusing on the linguistic and cultural contexts of Taiwan. 
Developed from a large base model, it's enriched with diverse Taiwanese textual sources and refined through Supervised Fine-Tuning. 
This model excels in language understanding and generation, aligning closely with Taiwan's cultural nuances. 
It demonstrates improved performance on various benchmarks like TC-Eval, showcasing its contextual comprehension and cultural relevance. 
For detailed insights into Taiwan LLM's development and features, refer to our [technical report](https://github.com/MiuLab/Taiwan-LLaMa/blob/main/twllm_paper.pdf).


## Model description

- **Model type:** A 7B parameter GPT-like model fine-tuned on a mix of publicly available, synthetic datasets.
- **Language(s) (NLP):** Primarily Traditional Chinese (zh-tw)
- **Finetuned from model:** [meta-llama/Llama-2-7b-hf](https://huggingface.co/yentinglin/meta-llama/Llama-2-7b-hf)

### Model Sources

<!-- Provide the basic links for the model. -->

- **Repository:** https://github.com/MiuLab/Taiwan-LLaMa
- **Demo:** https://twllm.com/

## Performance


![image/png](https://cdn-uploads.huggingface.co/production/uploads/5df9c78eda6d0311fd3d541f/HTwIzw6RDha2-PhuWqSuI.png)

## Intended uses

You should fine-tuned this model for instruction-following / chat application.

### Training hyperparameters

![image/png](https://cdn-uploads.huggingface.co/production/uploads/5df9c78eda6d0311fd3d541f/MdvHwdUvH-c926qyRAw7K.png)

![image/png](https://cdn-uploads.huggingface.co/production/uploads/5df9c78eda6d0311fd3d541f/kKpkvxDzOEyiAoTqmzRYO.png)


![image/png](https://cdn-uploads.huggingface.co/production/uploads/5df9c78eda6d0311fd3d541f/FsnlJ_fkRxf7fn5RKZnjE.png)

The following hyperparameters were used during training:
- learning_rate: 5e-05
- distributed_type: multi-GPU
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- lr_scheduler_warmup_ratio: 0.03
- num_epochs: 5.0

## Citation

If you find Taiwan LLM is useful in your work, please cite it with:

```
@misc{lin2023taiwan,
      title={Taiwan LLM: Bridging the Linguistic Divide with a Culturally Aligned Language Model}, 
      author={Yen-Ting Lin and Yun-Nung Chen},
      year={2023},
      eprint={2311.17487},
      archivePrefix={arXiv},
      primaryClass={cs.CL}
}
```

# Acknowledgement

Taiwan LLM v2 is conducted in collaboration with [Ubitus K.K.](http://ubitus.net). Ubitus provides valuable compute resources for the project.