richarderkhov/haoranxu_-_alma-13b-r-gguf overview

{
  "metadata": {},
  "card_data": {
    "frontmatter": {},
    "hero_image_url": "",
    "summary": "We release six translation models presented in the paper: Model checkpoints are released at huggingface: |     Models    | Base Model Link | LoRA Link | |:-------------:|:---------------:|:---------:| |    ALMA-7B    |        haoranxu/ALMA-7B        |     -     | |  ALMA-7B-LoRA |        haoranxu/ALMA-7B-Pretrain        |     haoranxu/ALMA-7B-Pretrain-LoRA     | |  **ALMA-7B-R (NEW!)** |        haoranxu/ALMA-7B-R (LoRA merged)        |     -    | |    ALMA-13B   |        haoranxu/ALMA-13B        |     -     | | ALMA-13B-LoRA |        haoranxu/ALMA-13B-Pretrain        |     haoranxu/ALMA-13B-Pretrain-LoRA     | | **ALMA-13B-R (NEW!)** |        haoranxu/ALMA-13B-R (LoRA merged)        |    -   | **Note that ALMA-7B-Pretrain and ALMA-13B-Pretrain are NOT translation models. They only experience stage 1 monolingual fine-tuning (20B tokens for the 7B model and 12B tokens for the 13B model), and should be utilized in conjunction with their LoRA models.** Datasets used by ALMA and ALMA-R are also released at huggingface now (NEW!) |     Datasets    | Train / Validation| Test | |:-------------:|:---------------:|:---------:| |    Human-Written Parallel Data (ALMA)    |        train and validation        |     WMT'22    | |  Triplet Preference Data |        train        |   WMT'22 and WMT'23   | A quick start to use our best system (ALMA-13B-R) for translation. An example of translating \"我爱机器翻译。\" into English: `` import torch from transformers import AutoModelForCausalLM from transformers import AutoTokenizer # Load base model and LoRA weights model = AutoModelForCausalLM.from_pretrained(\"haoranxu/ALMA-13B-R\", torch_dtype=torch.float16, device_map=\"auto\") tokenizer = AutoTokenizer.from_pretrained(\"haoranxu/ALMA-13B-R\", padding_side='left') # Add the source sentence into the prompt template prompt=\"Translate this from Chinese to English:\\nChinese: 我爱机器翻译。\\nEnglish:\" input_ids = tokenizer(prompt, return_tensors=\"pt\", padding=True, max_length=40, truncation=True).input_ids.cuda() # Translation with torch.no_grad(): generated_ids = model.generate(input_ids=input_ids, num_beams=5, max_new_tokens=20, do_sample=True, temperature=0.6, top_p=0.9) outputs = tokenizer.batch_decode(generated_ids, skip_special_tokens=True) print(outputs) `` Please find more details in our GitHub repository",
    "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\nALMA-13B-R - GGUF\n- Model creator: https://huggingface.co/haoranxu/\n- Original model: https://huggingface.co/haoranxu/ALMA-13B-R/\n\n\n| Name | Quant method | Size |\n| ---- | ---- | ---- |\n| [ALMA-13B-R.Q2_K.gguf](https://huggingface.co/RichardErkhov/haoranxu_-_ALMA-13B-R-gguf/blob/main/ALMA-13B-R.Q2_K.gguf) | Q2_K | 4.52GB |\n| [ALMA-13B-R.IQ3_XS.gguf](https://huggingface.co/RichardErkhov/haoranxu_-_ALMA-13B-R-gguf/blob/main/ALMA-13B-R.IQ3_XS.gguf) | IQ3_XS | 4.99GB |\n| [ALMA-13B-R.IQ3_S.gguf](https://huggingface.co/RichardErkhov/haoranxu_-_ALMA-13B-R-gguf/blob/main/ALMA-13B-R.IQ3_S.gguf) | IQ3_S | 5.27GB |\n| [ALMA-13B-R.Q3_K_S.gguf](https://huggingface.co/RichardErkhov/haoranxu_-_ALMA-13B-R-gguf/blob/main/ALMA-13B-R.Q3_K_S.gguf) | Q3_K_S | 5.27GB |\n| [ALMA-13B-R.IQ3_M.gguf](https://huggingface.co/RichardErkhov/haoranxu_-_ALMA-13B-R-gguf/blob/main/ALMA-13B-R.IQ3_M.gguf) | IQ3_M | 5.57GB |\n| [ALMA-13B-R.Q3_K.gguf](https://huggingface.co/RichardErkhov/haoranxu_-_ALMA-13B-R-gguf/blob/main/ALMA-13B-R.Q3_K.gguf) | Q3_K | 5.9GB |\n| [ALMA-13B-R.Q3_K_M.gguf](https://huggingface.co/RichardErkhov/haoranxu_-_ALMA-13B-R-gguf/blob/main/ALMA-13B-R.Q3_K_M.gguf) | Q3_K_M | 5.9GB |\n| [ALMA-13B-R.Q3_K_L.gguf](https://huggingface.co/RichardErkhov/haoranxu_-_ALMA-13B-R-gguf/blob/main/ALMA-13B-R.Q3_K_L.gguf) | Q3_K_L | 6.45GB |\n| [ALMA-13B-R.IQ4_XS.gguf](https://huggingface.co/RichardErkhov/haoranxu_-_ALMA-13B-R-gguf/blob/main/ALMA-13B-R.IQ4_XS.gguf) | IQ4_XS | 6.54GB |\n| [ALMA-13B-R.Q4_0.gguf](https://huggingface.co/RichardErkhov/haoranxu_-_ALMA-13B-R-gguf/blob/main/ALMA-13B-R.Q4_0.gguf) | Q4_0 | 6.86GB |\n| [ALMA-13B-R.IQ4_NL.gguf](https://huggingface.co/RichardErkhov/haoranxu_-_ALMA-13B-R-gguf/blob/main/ALMA-13B-R.IQ4_NL.gguf) | IQ4_NL | 6.9GB |\n| [ALMA-13B-R.Q4_K_S.gguf](https://huggingface.co/RichardErkhov/haoranxu_-_ALMA-13B-R-gguf/blob/main/ALMA-13B-R.Q4_K_S.gguf) | Q4_K_S | 6.91GB |\n| [ALMA-13B-R.Q4_K.gguf](https://huggingface.co/RichardErkhov/haoranxu_-_ALMA-13B-R-gguf/blob/main/ALMA-13B-R.Q4_K.gguf) | Q4_K | 7.33GB |\n| [ALMA-13B-R.Q4_K_M.gguf](https://huggingface.co/RichardErkhov/haoranxu_-_ALMA-13B-R-gguf/blob/main/ALMA-13B-R.Q4_K_M.gguf) | Q4_K_M | 7.33GB |\n| [ALMA-13B-R.Q4_1.gguf](https://huggingface.co/RichardErkhov/haoranxu_-_ALMA-13B-R-gguf/blob/main/ALMA-13B-R.Q4_1.gguf) | Q4_1 | 7.61GB |\n| [ALMA-13B-R.Q5_0.gguf](https://huggingface.co/RichardErkhov/haoranxu_-_ALMA-13B-R-gguf/blob/main/ALMA-13B-R.Q5_0.gguf) | Q5_0 | 8.36GB |\n| [ALMA-13B-R.Q5_K_S.gguf](https://huggingface.co/RichardErkhov/haoranxu_-_ALMA-13B-R-gguf/blob/main/ALMA-13B-R.Q5_K_S.gguf) | Q5_K_S | 8.36GB |\n| [ALMA-13B-R.Q5_K.gguf](https://huggingface.co/RichardErkhov/haoranxu_-_ALMA-13B-R-gguf/blob/main/ALMA-13B-R.Q5_K.gguf) | Q5_K | 8.6GB |\n| [ALMA-13B-R.Q5_K_M.gguf](https://huggingface.co/RichardErkhov/haoranxu_-_ALMA-13B-R-gguf/blob/main/ALMA-13B-R.Q5_K_M.gguf) | Q5_K_M | 8.6GB |\n| [ALMA-13B-R.Q5_1.gguf](https://huggingface.co/RichardErkhov/haoranxu_-_ALMA-13B-R-gguf/blob/main/ALMA-13B-R.Q5_1.gguf) | Q5_1 | 9.1GB |\n| [ALMA-13B-R.Q6_K.gguf](https://huggingface.co/RichardErkhov/haoranxu_-_ALMA-13B-R-gguf/blob/main/ALMA-13B-R.Q6_K.gguf) | Q6_K | 9.95GB |\n| [ALMA-13B-R.Q8_0.gguf](https://huggingface.co/RichardErkhov/haoranxu_-_ALMA-13B-R-gguf/blob/main/ALMA-13B-R.Q8_0.gguf) | Q8_0 | 12.88GB |\n\n\n\n\nOriginal model description:\n---\nlicense: mit\n---\n**[ALMA-R](https://arxiv.org/abs/2401.08417)** builds upon [ALMA models](https://arxiv.org/abs/2309.11674), with further LoRA fine-tuning with our proposed **Contrastive Preference Optimization (CPO)** as opposed to the Supervised Fine-tuning used in ALMA. CPO fine-tuning requires our [triplet preference data](https://huggingface.co/datasets/haoranxu/ALMA-R-Preference) for preference learning. ALMA-R now can matches or even exceeds GPT-4 or WMT winners!\n```\n@misc{xu2024contrastive,\n      title={Contrastive Preference Optimization: Pushing the Boundaries of LLM Performance in Machine Translation}, \n      author={Haoran Xu and Amr Sharaf and Yunmo Chen and Weiting Tan and Lingfeng Shen and Benjamin Van Durme and Kenton Murray and Young Jin Kim},\n      year={2024},\n      eprint={2401.08417},\n      archivePrefix={arXiv},\n      primaryClass={cs.CL}\n}\n```\n```\n@misc{xu2023paradigm,\n      title={A Paradigm Shift in Machine Translation: Boosting Translation Performance of Large Language Models}, \n      author={Haoran Xu and Young Jin Kim and Amr Sharaf and Hany Hassan Awadalla},\n      year={2023},\n      eprint={2309.11674},\n      archivePrefix={arXiv},\n      primaryClass={cs.CL}\n}\n```\n# Download ALMA(-R) Models and Dataset 🚀\n\nWe release six translation models presented in the paper:\n- ALMA-7B\n- ALMA-7B-LoRA\n- **ALMA-7B-R (NEW!)**: Further LoRA fine-tuning upon ALMA-7B-LoRA with contrastive preference optimization.\n- ALMA-13B\n- ALMA-13B-LoRA\n- **ALMA-13B-R (NEW!)**: Further LoRA fine-tuning upon ALMA-13B-LoRA with contrastive preference optimization (BEST MODEL!). \n  \nModel checkpoints are released at huggingface:\n|     Models    | Base Model Link | LoRA Link |\n|:-------------:|:---------------:|:---------:|\n|    ALMA-7B    |        [haoranxu/ALMA-7B](https://huggingface.co/haoranxu/ALMA-7B)        |     -     |\n|  ALMA-7B-LoRA |        [haoranxu/ALMA-7B-Pretrain](https://huggingface.co/haoranxu/ALMA-7B-Pretrain)        |     [haoranxu/ALMA-7B-Pretrain-LoRA](https://huggingface.co/haoranxu/ALMA-7B-Pretrain-LoRA)     |\n|  **ALMA-7B-R (NEW!)** |        [haoranxu/ALMA-7B-R (LoRA merged)](https://huggingface.co/haoranxu/ALMA-7B-R)        |     -    |\n|    ALMA-13B   |        [haoranxu/ALMA-13B](https://huggingface.co/haoranxu/ALMA-13B)        |     -     |\n| ALMA-13B-LoRA |        [haoranxu/ALMA-13B-Pretrain](https://huggingface.co/haoranxu/ALMA-13B-Pretrain)        |     [haoranxu/ALMA-13B-Pretrain-LoRA](https://huggingface.co/haoranxu/ALMA-13B-Pretrain-LoRA)     |\n| **ALMA-13B-R (NEW!)** |        [haoranxu/ALMA-13B-R (LoRA merged)](https://huggingface.co/haoranxu/ALMA-13B-R)        |    -   |\n\n**Note that `ALMA-7B-Pretrain` and `ALMA-13B-Pretrain` are NOT translation models. They only experience stage 1 monolingual fine-tuning (20B tokens for the 7B model and 12B tokens for the 13B model), and should be utilized in conjunction with their LoRA models.** \n\nDatasets used by ALMA and ALMA-R are also released at huggingface now (NEW!)\n|     Datasets    | Train / Validation| Test |\n|:-------------:|:---------------:|:---------:|\n|    Human-Written Parallel Data (ALMA)    |        [train and validation](https://huggingface.co/datasets/haoranxu/ALMA-Human-Parallel)        |     [WMT'22](https://huggingface.co/datasets/haoranxu/WMT22-Test)    |\n|  Triplet Preference Data |        [train](https://huggingface.co/datasets/haoranxu/ALMA-R-Preference)        |   [WMT'22](https://huggingface.co/datasets/haoranxu/WMT22-Test) and [WMT'23](https://huggingface.co/datasets/haoranxu/WMT23-Test)   |\n\n\nA quick start to use our best system (ALMA-13B-R) for translation. An example of translating \"我爱机器翻译。\" into English:\n```\nimport torch\nfrom transformers import AutoModelForCausalLM\nfrom transformers import AutoTokenizer\n\n# Load base model and LoRA weights\nmodel = AutoModelForCausalLM.from_pretrained(\"haoranxu/ALMA-13B-R\", torch_dtype=torch.float16, device_map=\"auto\")\ntokenizer = AutoTokenizer.from_pretrained(\"haoranxu/ALMA-13B-R\", padding_side='left')\n\n# Add the source sentence into the prompt template\nprompt=\"Translate this from Chinese to English:\\nChinese: 我爱机器翻译。\\nEnglish:\"\ninput_ids = tokenizer(prompt, return_tensors=\"pt\", padding=True, max_length=40, truncation=True).input_ids.cuda()\n\n# Translation\nwith torch.no_grad():\n    generated_ids = model.generate(input_ids=input_ids, num_beams=5, max_new_tokens=20, do_sample=True, temperature=0.6, top_p=0.9)\noutputs = tokenizer.batch_decode(generated_ids, skip_special_tokens=True)\nprint(outputs)\n```\n\nPlease find more details in our [GitHub repository](https://github.com/fe1ixxu/ALMA)\n\n",
    "related_quantizations": []
  },
  "tags": [
    "gguf",
    "arxiv:2401.08417",
    "arxiv:2309.11674",
    "endpoints_compatible",
    "region:us"
  ],
  "likes": 0,
  "downloads": 167,
  "gated": false,
  "private": false,
  "last_modified": "2024-05-19T15:00:20.000Z",
  "created_at": "2024-05-19T11:19:23.000Z",
  "pipeline_tag": "",
  "library_name": ""
}

{
  "_id": "6649e03bf604081903952ff8",
  "id": "RichardErkhov/haoranxu_-_ALMA-13B-R-gguf",
  "modelId": "RichardErkhov/haoranxu_-_ALMA-13B-R-gguf",
  "sha": "b38ecb443039bb35a59cd2718ce786b0ac338307",
  "createdAt": "2024-05-19T11:19:23.000Z",
  "lastModified": "2024-05-19T15:00:20.000Z",
  "author": "RichardErkhov",
  "downloads": 167,
  "likes": 0,
  "gated": false,
  "private": false,
  "pipeline_tag": "",
  "library_name": "",
  "siblings_count": 24
}