richarderkhov/hplt_-_sft-fpft-fr-bloom-1b1-gguf overview

{
  "metadata": {},
  "card_data": {
    "frontmatter": {},
    "hero_image_url": "",
    "summary": "Quantization made by Richard Erkhov. Github Discord Request more models sft-fpft-fr-bloom-1b1 - GGUF | Name | Quant method | Size | | ---- | ---- | ---- | | sft-fpft-fr-bloom-1b1.Q2_K.gguf | Q2_K | 0.66GB | | sft-fpft-fr-bloom-1b1.IQ3_XS.gguf | IQ3_XS | 0.73GB | | sft-fpft-fr-bloom-1b1.IQ3_S.gguf | IQ3_S | 0.73GB | | sft-fpft-fr-bloom-1b1.Q3_K_S.gguf | Q3_K_S | 0.73GB | | sft-fpft-fr-bloom-1b1.IQ3_M.gguf | IQ3_M | 0.77GB | | sft-fpft-fr-bloom-1b1.Q3_K.gguf | Q3_K | 0.79GB | | sft-fpft-fr-bloom-1b1.Q3_K_M.gguf | Q3_K_M | 0.79GB | | sft-fpft-fr-bloom-1b1.Q3_K_L.gguf | Q3_K_L | 0.82GB | | sft-fpft-fr-bloom-1b1.IQ4_XS.gguf | IQ4_XS | 0.84GB | | sft-fpft-fr-bloom-1b1.Q4_0.gguf | Q4_0 | 0.87GB | | sft-fpft-fr-bloom-1b1.IQ4_NL.gguf | IQ4_NL | 0.87GB | | sft-fpft-fr-bloom-1b1.Q4_K_S.gguf | Q4_K_S | 0.87GB | | sft-fpft-fr-bloom-1b1.Q4_K.gguf | Q4_K | 0.91GB | | sft-fpft-fr-bloom-1b1.Q4_K_M.gguf | Q4_K_M | 0.91GB | | sft-fpft-fr-bloom-1b1.Q4_1.gguf | Q4_1 | 0.93GB | | sft-fpft-fr-bloom-1b1.Q5_0.gguf | Q5_0 | 0.99GB | | sft-fpft-fr-bloom-1b1.Q5_K_S.gguf | Q5_K_S | 0.99GB | | sft-fpft-fr-bloom-1b1.Q5_K.gguf | Q5_K | 1.02GB | | sft-fpft-fr-bloom-1b1.Q5_K_M.gguf | Q5_K_M | 1.02GB | | sft-fpft-fr-bloom-1b1.Q5_1.gguf | Q5_1 | 1.05GB | | sft-fpft-fr-bloom-1b1.Q6_K.gguf | Q6_K | 1.12GB | | sft-fpft-fr-bloom-1b1.Q8_0.gguf | Q8_0 | 1.45GB | Original model description: --- language: tags: license: cc-by-nc-4.0 --- ###  Model Description This HF repository contains base LLMs instruction tuned (SFT) with full-parameter fine-tuning and then used to study whether monolingual or multilingual instruction tuning is more favourable. * GitHub * Paper #### Instruction tuning details * Base model: bloom-1b1 * Instruction tuning language: French * Training method: full-parameter fine-tuning. * Best checkpoint: best cross-entropy on a validation set, trained for 3 epochs. * Dataset: machine-translated from yahma/alpaca-cleaned. You can download our data HERE. #### Usage The model checkpoint should be loaded using transformers library. Please refer to our Github repository HERE for inference and training instructions. #### Citation `` @inproceedings{chen-etal-2024-monolingual, title=\"Monolingual or multilingual instruction tuning: Which makes a better {Alpaca}\", author=\"Pinzhen Chen and Shaoxiong Ji and Nikolay Bogoychev and Andrey Kutuzov and Barry Haddow and Kenneth Heafield\", year=\"2024\", booktitle = \"Findings of the Association for Computational Linguistics: EACL 2024\", } ``",
    "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\nsft-fpft-fr-bloom-1b1 - GGUF\n- Model creator: https://huggingface.co/HPLT/\n- Original model: https://huggingface.co/HPLT/sft-fpft-fr-bloom-1b1/\n\n\n| Name | Quant method | Size |\n| ---- | ---- | ---- |\n| [sft-fpft-fr-bloom-1b1.Q2_K.gguf](https://huggingface.co/RichardErkhov/HPLT_-_sft-fpft-fr-bloom-1b1-gguf/blob/main/sft-fpft-fr-bloom-1b1.Q2_K.gguf) | Q2_K | 0.66GB |\n| [sft-fpft-fr-bloom-1b1.IQ3_XS.gguf](https://huggingface.co/RichardErkhov/HPLT_-_sft-fpft-fr-bloom-1b1-gguf/blob/main/sft-fpft-fr-bloom-1b1.IQ3_XS.gguf) | IQ3_XS | 0.73GB |\n| [sft-fpft-fr-bloom-1b1.IQ3_S.gguf](https://huggingface.co/RichardErkhov/HPLT_-_sft-fpft-fr-bloom-1b1-gguf/blob/main/sft-fpft-fr-bloom-1b1.IQ3_S.gguf) | IQ3_S | 0.73GB |\n| [sft-fpft-fr-bloom-1b1.Q3_K_S.gguf](https://huggingface.co/RichardErkhov/HPLT_-_sft-fpft-fr-bloom-1b1-gguf/blob/main/sft-fpft-fr-bloom-1b1.Q3_K_S.gguf) | Q3_K_S | 0.73GB |\n| [sft-fpft-fr-bloom-1b1.IQ3_M.gguf](https://huggingface.co/RichardErkhov/HPLT_-_sft-fpft-fr-bloom-1b1-gguf/blob/main/sft-fpft-fr-bloom-1b1.IQ3_M.gguf) | IQ3_M | 0.77GB |\n| [sft-fpft-fr-bloom-1b1.Q3_K.gguf](https://huggingface.co/RichardErkhov/HPLT_-_sft-fpft-fr-bloom-1b1-gguf/blob/main/sft-fpft-fr-bloom-1b1.Q3_K.gguf) | Q3_K | 0.79GB |\n| [sft-fpft-fr-bloom-1b1.Q3_K_M.gguf](https://huggingface.co/RichardErkhov/HPLT_-_sft-fpft-fr-bloom-1b1-gguf/blob/main/sft-fpft-fr-bloom-1b1.Q3_K_M.gguf) | Q3_K_M | 0.79GB |\n| [sft-fpft-fr-bloom-1b1.Q3_K_L.gguf](https://huggingface.co/RichardErkhov/HPLT_-_sft-fpft-fr-bloom-1b1-gguf/blob/main/sft-fpft-fr-bloom-1b1.Q3_K_L.gguf) | Q3_K_L | 0.82GB |\n| [sft-fpft-fr-bloom-1b1.IQ4_XS.gguf](https://huggingface.co/RichardErkhov/HPLT_-_sft-fpft-fr-bloom-1b1-gguf/blob/main/sft-fpft-fr-bloom-1b1.IQ4_XS.gguf) | IQ4_XS | 0.84GB |\n| [sft-fpft-fr-bloom-1b1.Q4_0.gguf](https://huggingface.co/RichardErkhov/HPLT_-_sft-fpft-fr-bloom-1b1-gguf/blob/main/sft-fpft-fr-bloom-1b1.Q4_0.gguf) | Q4_0 | 0.87GB |\n| [sft-fpft-fr-bloom-1b1.IQ4_NL.gguf](https://huggingface.co/RichardErkhov/HPLT_-_sft-fpft-fr-bloom-1b1-gguf/blob/main/sft-fpft-fr-bloom-1b1.IQ4_NL.gguf) | IQ4_NL | 0.87GB |\n| [sft-fpft-fr-bloom-1b1.Q4_K_S.gguf](https://huggingface.co/RichardErkhov/HPLT_-_sft-fpft-fr-bloom-1b1-gguf/blob/main/sft-fpft-fr-bloom-1b1.Q4_K_S.gguf) | Q4_K_S | 0.87GB |\n| [sft-fpft-fr-bloom-1b1.Q4_K.gguf](https://huggingface.co/RichardErkhov/HPLT_-_sft-fpft-fr-bloom-1b1-gguf/blob/main/sft-fpft-fr-bloom-1b1.Q4_K.gguf) | Q4_K | 0.91GB |\n| [sft-fpft-fr-bloom-1b1.Q4_K_M.gguf](https://huggingface.co/RichardErkhov/HPLT_-_sft-fpft-fr-bloom-1b1-gguf/blob/main/sft-fpft-fr-bloom-1b1.Q4_K_M.gguf) | Q4_K_M | 0.91GB |\n| [sft-fpft-fr-bloom-1b1.Q4_1.gguf](https://huggingface.co/RichardErkhov/HPLT_-_sft-fpft-fr-bloom-1b1-gguf/blob/main/sft-fpft-fr-bloom-1b1.Q4_1.gguf) | Q4_1 | 0.93GB |\n| [sft-fpft-fr-bloom-1b1.Q5_0.gguf](https://huggingface.co/RichardErkhov/HPLT_-_sft-fpft-fr-bloom-1b1-gguf/blob/main/sft-fpft-fr-bloom-1b1.Q5_0.gguf) | Q5_0 | 0.99GB |\n| [sft-fpft-fr-bloom-1b1.Q5_K_S.gguf](https://huggingface.co/RichardErkhov/HPLT_-_sft-fpft-fr-bloom-1b1-gguf/blob/main/sft-fpft-fr-bloom-1b1.Q5_K_S.gguf) | Q5_K_S | 0.99GB |\n| [sft-fpft-fr-bloom-1b1.Q5_K.gguf](https://huggingface.co/RichardErkhov/HPLT_-_sft-fpft-fr-bloom-1b1-gguf/blob/main/sft-fpft-fr-bloom-1b1.Q5_K.gguf) | Q5_K | 1.02GB |\n| [sft-fpft-fr-bloom-1b1.Q5_K_M.gguf](https://huggingface.co/RichardErkhov/HPLT_-_sft-fpft-fr-bloom-1b1-gguf/blob/main/sft-fpft-fr-bloom-1b1.Q5_K_M.gguf) | Q5_K_M | 1.02GB |\n| [sft-fpft-fr-bloom-1b1.Q5_1.gguf](https://huggingface.co/RichardErkhov/HPLT_-_sft-fpft-fr-bloom-1b1-gguf/blob/main/sft-fpft-fr-bloom-1b1.Q5_1.gguf) | Q5_1 | 1.05GB |\n| [sft-fpft-fr-bloom-1b1.Q6_K.gguf](https://huggingface.co/RichardErkhov/HPLT_-_sft-fpft-fr-bloom-1b1-gguf/blob/main/sft-fpft-fr-bloom-1b1.Q6_K.gguf) | Q6_K | 1.12GB |\n| [sft-fpft-fr-bloom-1b1.Q8_0.gguf](https://huggingface.co/RichardErkhov/HPLT_-_sft-fpft-fr-bloom-1b1-gguf/blob/main/sft-fpft-fr-bloom-1b1.Q8_0.gguf) | Q8_0 | 1.45GB |\n\n\n\n\nOriginal model description:\n\n---\nlanguage:\n- fr\ntags:\n- generation\n- question answering\n- instruction tuning\nlicense: cc-by-nc-4.0\n---\n\n###  Model Description\n\nThis HF repository contains base LLMs instruction tuned (SFT) with full-parameter fine-tuning and then used to study whether monolingual or multilingual instruction tuning is more favourable.\n* [GitHub](https://github.com/hplt-project/monolingual-multilingual-instruction-tuning/tree/main)\n* [Paper](https://arxiv.org/abs/2309.08958)\n\n#### Instruction tuning details\n* Base model: [bloom-1b1](https://huggingface.co/bloom-1b1)\n* Instruction tuning language: French\n* Training method: full-parameter fine-tuning.\n* Best checkpoint: best cross-entropy on a validation set, trained for 3 epochs.\n* Dataset: machine-translated from [yahma/alpaca-cleaned](https://huggingface.co/datasets/yahma/alpaca-cleaned). You can download our data [HERE](https://github.com/hplt-project/monolingual-multilingual-instruction-tuning/tree/main/training-data).\n\n#### Usage\nThe model checkpoint should be loaded using `transformers` library.\n\nPlease refer to our Github repository [HERE](https://github.com/hplt-project/monolingual-multilingual-instruction-tuning/tree/main/fpft) for inference and training instructions.\n\n#### Citation\n```\n@inproceedings{chen-etal-2024-monolingual,\n  title=\"Monolingual or multilingual instruction tuning: Which makes a better {Alpaca}\",\n  author=\"Pinzhen Chen and Shaoxiong Ji and Nikolay Bogoychev and Andrey Kutuzov and Barry Haddow and Kenneth Heafield\",\n  year=\"2024\",\n  booktitle = \"Findings of the Association for Computational Linguistics: EACL 2024\",\n}\n```\n\n\n\n",
    "related_quantizations": []
  },
  "tags": [
    "gguf",
    "arxiv:2309.08958",
    "endpoints_compatible",
    "region:us"
  ],
  "likes": 0,
  "downloads": 154,
  "gated": false,
  "private": false,
  "last_modified": "2025-04-04T00:12:43.000Z",
  "created_at": "2025-04-03T23:22:35.000Z",
  "pipeline_tag": "",
  "library_name": ""
}

{
  "_id": "67ef183b72af360e0358b510",
  "id": "RichardErkhov/HPLT_-_sft-fpft-fr-bloom-1b1-gguf",
  "modelId": "RichardErkhov/HPLT_-_sft-fpft-fr-bloom-1b1-gguf",
  "sha": "4188a209e54fa6c488d4aa952074cf8324dea2a6",
  "createdAt": "2025-04-03T23:22:35.000Z",
  "lastModified": "2025-04-04T00:12:43.000Z",
  "author": "RichardErkhov",
  "downloads": 154,
  "likes": 0,
  "gated": false,
  "private": false,
  "pipeline_tag": "",
  "library_name": "",
  "siblings_count": 24
}