richarderkhov/lzw1008_-_emollama-chat-13b-gguf overview

{
  "metadata": {},
  "card_data": {
    "frontmatter": {},
    "hero_image_url": "",
    "summary": "Emollama-chat-13b is part of the EmoLLMs project, the first open-source large language model (LLM) series for comprehensive affective analysis with instruction-following capability. This model is finetuned based on the Meta LLaMA2-chat-13B foundation model and the full AAID instruction tuning data. The model can be used for affective classification tasks (e.g. sentimental polarity or categorical emotions), and regression tasks (e.g. sentiment strength or emotion intensity). # Ethical Consideration Recent studies have indicated LLMs may introduce some potential bias, such as gender gaps. Meanwhile, some incorrect prediction results, and over-generalization also illustrate the potential risks of current LLMs. Therefore, there are still many challenges in applying the model to real-scenario affective analysis systems.",
    "quick_links": [],
    "benchmark_table_html": "",
    "readme_markdown": "Quantization made by Richard Erkhov.\n\n[Github](https://github.com/RichardErkhov)\n\n[Discord](https://discord.gg/pvy7H8DZMG)\n\n[Request more models](https://github.com/RichardErkhov/quant_request)\n\n\nEmollama-chat-13b - GGUF\n- Model creator: https://huggingface.co/lzw1008/\n- Original model: https://huggingface.co/lzw1008/Emollama-chat-13b/\n\n\n| Name | Quant method | Size |\n| ---- | ---- | ---- |\n| [Emollama-chat-13b.Q2_K.gguf](https://huggingface.co/RichardErkhov/lzw1008_-_Emollama-chat-13b-gguf/blob/main/Emollama-chat-13b.Q2_K.gguf) | Q2_K | 4.52GB |\n| [Emollama-chat-13b.IQ3_XS.gguf](https://huggingface.co/RichardErkhov/lzw1008_-_Emollama-chat-13b-gguf/blob/main/Emollama-chat-13b.IQ3_XS.gguf) | IQ3_XS | 4.99GB |\n| [Emollama-chat-13b.IQ3_S.gguf](https://huggingface.co/RichardErkhov/lzw1008_-_Emollama-chat-13b-gguf/blob/main/Emollama-chat-13b.IQ3_S.gguf) | IQ3_S | 5.27GB |\n| [Emollama-chat-13b.Q3_K_S.gguf](https://huggingface.co/RichardErkhov/lzw1008_-_Emollama-chat-13b-gguf/blob/main/Emollama-chat-13b.Q3_K_S.gguf) | Q3_K_S | 5.27GB |\n| [Emollama-chat-13b.IQ3_M.gguf](https://huggingface.co/RichardErkhov/lzw1008_-_Emollama-chat-13b-gguf/blob/main/Emollama-chat-13b.IQ3_M.gguf) | IQ3_M | 5.57GB |\n| [Emollama-chat-13b.Q3_K.gguf](https://huggingface.co/RichardErkhov/lzw1008_-_Emollama-chat-13b-gguf/blob/main/Emollama-chat-13b.Q3_K.gguf) | Q3_K | 5.9GB |\n| [Emollama-chat-13b.Q3_K_M.gguf](https://huggingface.co/RichardErkhov/lzw1008_-_Emollama-chat-13b-gguf/blob/main/Emollama-chat-13b.Q3_K_M.gguf) | Q3_K_M | 5.9GB |\n| [Emollama-chat-13b.Q3_K_L.gguf](https://huggingface.co/RichardErkhov/lzw1008_-_Emollama-chat-13b-gguf/blob/main/Emollama-chat-13b.Q3_K_L.gguf) | Q3_K_L | 6.45GB |\n| [Emollama-chat-13b.IQ4_XS.gguf](https://huggingface.co/RichardErkhov/lzw1008_-_Emollama-chat-13b-gguf/blob/main/Emollama-chat-13b.IQ4_XS.gguf) | IQ4_XS | 6.54GB |\n| [Emollama-chat-13b.Q4_0.gguf](https://huggingface.co/RichardErkhov/lzw1008_-_Emollama-chat-13b-gguf/blob/main/Emollama-chat-13b.Q4_0.gguf) | Q4_0 | 6.86GB |\n| [Emollama-chat-13b.IQ4_NL.gguf](https://huggingface.co/RichardErkhov/lzw1008_-_Emollama-chat-13b-gguf/blob/main/Emollama-chat-13b.IQ4_NL.gguf) | IQ4_NL | 6.9GB |\n| [Emollama-chat-13b.Q4_K_S.gguf](https://huggingface.co/RichardErkhov/lzw1008_-_Emollama-chat-13b-gguf/blob/main/Emollama-chat-13b.Q4_K_S.gguf) | Q4_K_S | 6.91GB |\n| [Emollama-chat-13b.Q4_K.gguf](https://huggingface.co/RichardErkhov/lzw1008_-_Emollama-chat-13b-gguf/blob/main/Emollama-chat-13b.Q4_K.gguf) | Q4_K | 7.33GB |\n| [Emollama-chat-13b.Q4_K_M.gguf](https://huggingface.co/RichardErkhov/lzw1008_-_Emollama-chat-13b-gguf/blob/main/Emollama-chat-13b.Q4_K_M.gguf) | Q4_K_M | 7.33GB |\n| [Emollama-chat-13b.Q4_1.gguf](https://huggingface.co/RichardErkhov/lzw1008_-_Emollama-chat-13b-gguf/blob/main/Emollama-chat-13b.Q4_1.gguf) | Q4_1 | 7.61GB |\n| [Emollama-chat-13b.Q5_0.gguf](https://huggingface.co/RichardErkhov/lzw1008_-_Emollama-chat-13b-gguf/blob/main/Emollama-chat-13b.Q5_0.gguf) | Q5_0 | 8.36GB |\n| [Emollama-chat-13b.Q5_K_S.gguf](https://huggingface.co/RichardErkhov/lzw1008_-_Emollama-chat-13b-gguf/blob/main/Emollama-chat-13b.Q5_K_S.gguf) | Q5_K_S | 8.36GB |\n| [Emollama-chat-13b.Q5_K.gguf](https://huggingface.co/RichardErkhov/lzw1008_-_Emollama-chat-13b-gguf/blob/main/Emollama-chat-13b.Q5_K.gguf) | Q5_K | 8.6GB |\n| [Emollama-chat-13b.Q5_K_M.gguf](https://huggingface.co/RichardErkhov/lzw1008_-_Emollama-chat-13b-gguf/blob/main/Emollama-chat-13b.Q5_K_M.gguf) | Q5_K_M | 8.6GB |\n| [Emollama-chat-13b.Q5_1.gguf](https://huggingface.co/RichardErkhov/lzw1008_-_Emollama-chat-13b-gguf/blob/main/Emollama-chat-13b.Q5_1.gguf) | Q5_1 | 9.1GB |\n| [Emollama-chat-13b.Q6_K.gguf](https://huggingface.co/RichardErkhov/lzw1008_-_Emollama-chat-13b-gguf/blob/main/Emollama-chat-13b.Q6_K.gguf) | Q6_K | 9.95GB |\n| [Emollama-chat-13b.Q8_0.gguf](https://huggingface.co/RichardErkhov/lzw1008_-_Emollama-chat-13b-gguf/blob/main/Emollama-chat-13b.Q8_0.gguf) | Q8_0 | 12.88GB |\n\n\n\n\nOriginal model description:\n---\nlicense: mit\nlanguage:\n- en\n---\n\n# Introduction\n\nEmollama-chat-13b is part of the [EmoLLMs](https://github.com/lzw108/EmoLLMs) project, the first open-source large language model (LLM) series for \ncomprehensive affective analysis with instruction-following capability. This model is finetuned based on the Meta LLaMA2-chat-13B foundation model and the full AAID instruction tuning data.\nThe model can be used for affective classification tasks (e.g. sentimental polarity\nor categorical emotions), and regression tasks (e.g. sentiment strength or emotion intensity).\n\n# Ethical Consideration\n\nRecent studies have indicated LLMs may introduce some potential\nbias, such as gender gaps. Meanwhile, some incorrect prediction results, and over-generalization\nalso illustrate the potential risks of current LLMs. Therefore, there\nare still many challenges in applying the model to real-scenario\naffective analysis systems.\n\n## Models in EmoLLMs\n\nThere are a series of EmoLLMs, including Emollama-7b, Emollama-chat-7b, Emollama-chat-13b,  Emoopt-13b, Emobloom-7b, Emot5-large, Emobart-large.\n\n- **Emollama-7b**: This model is finetuned based on the LLaMA2-7B. \n- **Emollama-chat-7b**: This model is finetuned based on the LLaMA2-chat-7B. \n- **Emollama-chat-13b**: This model is finetuned based on the LLaMA2-chat-13B. \n- **Emoopt-13b**: This model is finetuned based on the OPT-13B. \n- **Emobloom-7b**: This model is finetuned based on the Bloomz-7b1-mt. \n- **Emot5-large**: This model is finetuned based on the T5-large. \n- **Emobart-large**: This model is finetuned based on the bart-large.\n\nAll models are trained on the full AAID instruction tuning data.\n\n\n\n## Usage\n\nYou can use the Emollama-chat-13b model in your Python project with the Hugging Face Transformers library. Here is a simple example of how to load the model:\n\n```python\nfrom transformers import LlamaTokenizer, LlamaForCausalLM\ntokenizer = LlamaTokenizer.from_pretrained('lzw1008/Emollama-chat-13b')\nmodel = LlamaForCausalLM.from_pretrained('lzw1008/Emollama-chat-13b', device_map='auto')\n```\n\nIn this example, LlamaTokenizer is used to load the tokenizer, and LlamaForCausalLM is used to load the model. The `device_map='auto'` argument is used to automatically\nuse the GPU if it's available.\n\n## Prompt examples\n \n\n### Emotion intensity\n\n    Human: \n    Task: Assign a numerical value between 0 (least E) and 1 (most E) to represent the intensity of emotion E expressed in the text.\n    Text: @CScheiwiller can't stop smiling 😆😆😆\n    Emotion: joy\n    Intensity Score:\n\n    Assistant:\n    >>0.896\n\n### Sentiment strength\n\n    Human:\n    Task: Evaluate the valence intensity of the writer's mental state based on the text, assigning it a real-valued score from 0 (most negative) to 1 (most positive).\n    Text: Happy Birthday shorty. Stay fine stay breezy stay wavy @daviistuart 😘\n    Intensity Score:\n\n    Assistant:\n    >>0.879\n\n### Sentiment classification\n\n    Human:\n    Task: Categorize the text into an ordinal class that best characterizes the writer's mental state, considering various degrees of positive and negative sentiment intensity. 3: very positive mental state can be inferred. 2: moderately positive mental state can be inferred. 1: slightly positive mental state can be inferred. 0: neutral or mixed mental state can be inferred. -1: slightly negative mental state can be inferred. -2: moderately negative mental state can be inferred. -3: very negative mental state can be inferred\n    Text: Beyoncé resentment gets me in my feelings every time. 😩\n    Intensity Class:\n\n    Assistant:\n    >>-3: very negative emotional state can be inferred\n\n### Emotion classification\n\n    Human:\n    Task: Categorize the text's emotional tone as either 'neutral or no emotion' or identify the presence of one or more of the given emotions (anger, anticipation, disgust, fear, joy, love, optimism, pessimism, sadness, surprise, trust).\n    Text: Whatever you decide to do make sure it makes you #happy.\n    This text contains emotions:\n\n    Assistant:\n    >>joy, love, optimism\n\nThe task description can be adjusted according to the specific task.\n\n## License\n\nEmoLLMs series are licensed under MIT. For more details, please see the MIT file.\n\n## Citation\n\nIf you use the series of EmoLLMs in your work, please cite our paper:\n\n```bibtex\n@article{liu2024emollms,\n  title={EmoLLMs: A Series of Emotional Large Language Models and Annotation Tools for Comprehensive Affective Analysis},\n  author={Liu, Zhiwei and Yang, Kailai and Zhang, Tianlin and Xie, Qianqian and Yu, Zeping and Ananiadou, Sophia},\n  journal={arXiv preprint arXiv:2401.08508},\n  year={2024}\n}\n```\n\n",
    "related_quantizations": []
  },
  "tags": [
    "gguf",
    "arxiv:2401.08508",
    "endpoints_compatible",
    "region:us"
  ],
  "likes": 0,
  "downloads": 888,
  "gated": false,
  "private": false,
  "last_modified": "2024-08-06T18:19:13.000Z",
  "created_at": "2024-08-06T05:07:22.000Z",
  "pipeline_tag": "",
  "library_name": ""
}

{
  "_id": "66b1af8a418ff05dfe62cb96",
  "id": "RichardErkhov/lzw1008_-_Emollama-chat-13b-gguf",
  "modelId": "RichardErkhov/lzw1008_-_Emollama-chat-13b-gguf",
  "sha": "6b922bd589de770d55f7509ea016bfb69e76562d",
  "createdAt": "2024-08-06T05:07:22.000Z",
  "lastModified": "2024-08-06T18:19:13.000Z",
  "author": "RichardErkhov",
  "downloads": 888,
  "likes": 0,
  "gated": false,
  "private": false,
  "pipeline_tag": "",
  "library_name": "",
  "siblings_count": 24
}