RichardErkhov/yentinglin_-_Taiwan-LLM-7B-v2.1-chat-gguf

Quantization made by Richard Erkhov.

Github

Discord

Request more models

Taiwan-LLM-7B-v2.1-chat - GGUF

• Model creator: https://huggingface.co/yentinglin/
• Original model: https://huggingface.co/yentinglin/Taiwan-LLM-7B-v2.1-chat/

Name	Quant method	Size
Taiwan-LLM-7B-v2.1-chat.Q2_K.gguf	Q2_K	2.36GB
Taiwan-LLM-7B-v2.1-chat.IQ3_XS.gguf	IQ3_XS	2.6GB
Taiwan-LLM-7B-v2.1-chat.IQ3_S.gguf	IQ3_S	2.75GB
Taiwan-LLM-7B-v2.1-chat.Q3_K_S.gguf	Q3_K_S	2.75GB
Taiwan-LLM-7B-v2.1-chat.IQ3_M.gguf	IQ3_M	2.9GB
Taiwan-LLM-7B-v2.1-chat.Q3_K.gguf	Q3_K	3.07GB
Taiwan-LLM-7B-v2.1-chat.Q3_K_M.gguf	Q3_K_M	3.07GB
Taiwan-LLM-7B-v2.1-chat.Q3_K_L.gguf	Q3_K_L	3.35GB
Taiwan-LLM-7B-v2.1-chat.IQ4_XS.gguf	IQ4_XS	3.4GB
Taiwan-LLM-7B-v2.1-chat.Q4_0.gguf	Q4_0	3.56GB
Taiwan-LLM-7B-v2.1-chat.IQ4_NL.gguf	IQ4_NL	3.58GB
Taiwan-LLM-7B-v2.1-chat.Q4_K_S.gguf	Q4_K_S	3.59GB
Taiwan-LLM-7B-v2.1-chat.Q4_K.gguf	Q4_K	3.8GB
Taiwan-LLM-7B-v2.1-chat.Q4_K_M.gguf	Q4_K_M	3.8GB
Taiwan-LLM-7B-v2.1-chat.Q4_1.gguf	Q4_1	3.95GB
Taiwan-LLM-7B-v2.1-chat.Q5_0.gguf	Q5_0	4.33GB
Taiwan-LLM-7B-v2.1-chat.Q5_K_S.gguf	Q5_K_S	4.33GB
Taiwan-LLM-7B-v2.1-chat.Q5_K.gguf	Q5_K	4.45GB
Taiwan-LLM-7B-v2.1-chat.Q5_K_M.gguf	Q5_K_M	4.45GB
Taiwan-LLM-7B-v2.1-chat.Q5_1.gguf	Q5_1	4.72GB
Taiwan-LLM-7B-v2.1-chat.Q6_K.gguf	Q6_K	5.15GB

Original model description:

---

For reference on model card metadata, see the spec: https://github.com/huggingface/hub-docs/blob/main/modelcard.md?plain=1

Doc / guide: https://huggingface.co/docs/hub/model-cards

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

🌟 Checkout Taiwan-LLM Demo Chat-UI 🌟

Model Card for Taiwan LLM 7B v2.1 chat

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.

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: yentinglin/Taiwan-LLM-7B-v2.0-base
• TMMLUS+ score: 22.19570181818182

Model Sources

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

Performance

Intended uses

Here's how you can run the model using the pipeline() function from 🤗 Transformers:

# pip install transformers>=4.34
# pip install accelerate

import torch
from transformers import pipeline

pipe = pipeline("text-generation", model="yentinglin/Taiwan-LLM-7B-v2.1-chat", torch_dtype=torch.bfloat16, device_map="auto")

# We use the tokenizer's chat template to format each message - see https://huggingface.co/docs/transformers/main/en/chat_templating
messages = [
    {
        "role": "system",
        "content": "你是一個人工智慧助理",
    },
    {"role": "user", "content": "東北季風如何影響台灣氣候？"},
]
prompt = pipe.tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
outputs = pipe(prompt, max_new_tokens=256, do_sample=True, temperature=0.7, top_k=50, top_p=0.95)
print(outputs[0]["generated_text"])

Training hyperparameters

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.. Ubitus provides valuable compute resources for the project.