richarderkhov/oopere_-_martra-open-gemma-2b-it-dpo-gguf overview

{
  "metadata": {},
  "card_data": {
    "frontmatter": {},
    "hero_image_url": "",
    "summary": "Quantization made by Richard Erkhov. Github Discord Request more models martra-open-gemma-2b-it-dpo - GGUF | Name | Quant method | Size | | ---- | ---- | ---- | | martra-open-gemma-2b-it-dpo.Q2_K.gguf | Q2_K | 1.08GB | | martra-open-gemma-2b-it-dpo.IQ3_XS.gguf | IQ3_XS | 1.16GB | | martra-open-gemma-2b-it-dpo.IQ3_S.gguf | IQ3_S | 1.2GB | | martra-open-gemma-2b-it-dpo.Q3_K_S.gguf | Q3_K_S | 1.2GB | | martra-open-gemma-2b-it-dpo.IQ3_M.gguf | IQ3_M | 1.22GB | | martra-open-gemma-2b-it-dpo.Q3_K.gguf | Q3_K | 1.29GB | | martra-open-gemma-2b-it-dpo.Q3_K_M.gguf | Q3_K_M | 1.29GB | | martra-open-gemma-2b-it-dpo.Q3_K_L.gguf | Q3_K_L | 1.36GB | | martra-open-gemma-2b-it-dpo.IQ4_XS.gguf | IQ4_XS | 1.4GB | | martra-open-gemma-2b-it-dpo.Q4_0.gguf | Q4_0 | 1.44GB | | martra-open-gemma-2b-it-dpo.IQ4_NL.gguf | IQ4_NL | 1.45GB | | martra-open-gemma-2b-it-dpo.Q4_K_S.gguf | Q4_K_S | 1.45GB | | martra-open-gemma-2b-it-dpo.Q4_K.gguf | Q4_K | 1.52GB | | martra-open-gemma-2b-it-dpo.Q4_K_M.gguf | Q4_K_M | 1.52GB | | martra-open-gemma-2b-it-dpo.Q4_1.gguf | Q4_1 | 1.56GB | | martra-open-gemma-2b-it-dpo.Q5_0.gguf | Q5_0 | 1.68GB | | martra-open-gemma-2b-it-dpo.Q5_K_S.gguf | Q5_K_S | 1.68GB | | martra-open-gemma-2b-it-dpo.Q5_K.gguf | Q5_K | 1.71GB | | martra-open-gemma-2b-it-dpo.Q5_K_M.gguf | Q5_K_M | 1.71GB | | martra-open-gemma-2b-it-dpo.Q5_1.gguf | Q5_1 | 1.79GB | | martra-open-gemma-2b-it-dpo.Q6_K.gguf | Q6_K | 1.92GB | | martra-open-gemma-2b-it-dpo.Q8_0.gguf | Q8_0 | 2.49GB | Original model description: --- language: license: mit library_name: transformers tags: datasets: pipeline_tag: text-generation widget: example_title: 'Return only numbers. ' respectively. In how many ways can this be done? example_title: Solve Problem --- You can see the process with instructions for creating the model in the notebook: Aligning_DPO_gemma-2b.ipynb To create it, we started with the google/gemma-2b-it model and applied DPO alignment using the distilabel-capybara-dpo-7k-binarized dataset. Gemma is a family of lightweight, state-of-the-art open models from Google, built from the same research and technology used to create the Gemini models. **Row in the Dataset:** *{'prompt': '\\nAssist me in calculating 9319357631 plus 595. Numbers and symbols only, please.\\n\\n', 'chosen': 'The sum of 9319357631 and 595 is 9319358226.\\n', 'rejected': 'The result of adding 9319357631 and 595 is 9319363626.\\n'}* **Prompt:** *3713841893836/4? Limit your response to mathematical expressions and symbols.* **Response from the Base model:** *To find the result of the division, we can simply divide the given number by 4: $$ \\frac{3713841893836}{4} = 928460473459* **Response from the fine-tuned model:** *3713841893836 ÷ 4 = 928460473459* If you want to see how the model was created, you can check out the repository where the book's notebooks are kept up-to-date.",
    "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\nmartra-open-gemma-2b-it-dpo - GGUF\n- Model creator: https://huggingface.co/oopere/\n- Original model: https://huggingface.co/oopere/martra-open-gemma-2b-it-dpo/\n\n\n| Name | Quant method | Size |\n| ---- | ---- | ---- |\n| [martra-open-gemma-2b-it-dpo.Q2_K.gguf](https://huggingface.co/RichardErkhov/oopere_-_martra-open-gemma-2b-it-dpo-gguf/blob/main/martra-open-gemma-2b-it-dpo.Q2_K.gguf) | Q2_K | 1.08GB |\n| [martra-open-gemma-2b-it-dpo.IQ3_XS.gguf](https://huggingface.co/RichardErkhov/oopere_-_martra-open-gemma-2b-it-dpo-gguf/blob/main/martra-open-gemma-2b-it-dpo.IQ3_XS.gguf) | IQ3_XS | 1.16GB |\n| [martra-open-gemma-2b-it-dpo.IQ3_S.gguf](https://huggingface.co/RichardErkhov/oopere_-_martra-open-gemma-2b-it-dpo-gguf/blob/main/martra-open-gemma-2b-it-dpo.IQ3_S.gguf) | IQ3_S | 1.2GB |\n| [martra-open-gemma-2b-it-dpo.Q3_K_S.gguf](https://huggingface.co/RichardErkhov/oopere_-_martra-open-gemma-2b-it-dpo-gguf/blob/main/martra-open-gemma-2b-it-dpo.Q3_K_S.gguf) | Q3_K_S | 1.2GB |\n| [martra-open-gemma-2b-it-dpo.IQ3_M.gguf](https://huggingface.co/RichardErkhov/oopere_-_martra-open-gemma-2b-it-dpo-gguf/blob/main/martra-open-gemma-2b-it-dpo.IQ3_M.gguf) | IQ3_M | 1.22GB |\n| [martra-open-gemma-2b-it-dpo.Q3_K.gguf](https://huggingface.co/RichardErkhov/oopere_-_martra-open-gemma-2b-it-dpo-gguf/blob/main/martra-open-gemma-2b-it-dpo.Q3_K.gguf) | Q3_K | 1.29GB |\n| [martra-open-gemma-2b-it-dpo.Q3_K_M.gguf](https://huggingface.co/RichardErkhov/oopere_-_martra-open-gemma-2b-it-dpo-gguf/blob/main/martra-open-gemma-2b-it-dpo.Q3_K_M.gguf) | Q3_K_M | 1.29GB |\n| [martra-open-gemma-2b-it-dpo.Q3_K_L.gguf](https://huggingface.co/RichardErkhov/oopere_-_martra-open-gemma-2b-it-dpo-gguf/blob/main/martra-open-gemma-2b-it-dpo.Q3_K_L.gguf) | Q3_K_L | 1.36GB |\n| [martra-open-gemma-2b-it-dpo.IQ4_XS.gguf](https://huggingface.co/RichardErkhov/oopere_-_martra-open-gemma-2b-it-dpo-gguf/blob/main/martra-open-gemma-2b-it-dpo.IQ4_XS.gguf) | IQ4_XS | 1.4GB |\n| [martra-open-gemma-2b-it-dpo.Q4_0.gguf](https://huggingface.co/RichardErkhov/oopere_-_martra-open-gemma-2b-it-dpo-gguf/blob/main/martra-open-gemma-2b-it-dpo.Q4_0.gguf) | Q4_0 | 1.44GB |\n| [martra-open-gemma-2b-it-dpo.IQ4_NL.gguf](https://huggingface.co/RichardErkhov/oopere_-_martra-open-gemma-2b-it-dpo-gguf/blob/main/martra-open-gemma-2b-it-dpo.IQ4_NL.gguf) | IQ4_NL | 1.45GB |\n| [martra-open-gemma-2b-it-dpo.Q4_K_S.gguf](https://huggingface.co/RichardErkhov/oopere_-_martra-open-gemma-2b-it-dpo-gguf/blob/main/martra-open-gemma-2b-it-dpo.Q4_K_S.gguf) | Q4_K_S | 1.45GB |\n| [martra-open-gemma-2b-it-dpo.Q4_K.gguf](https://huggingface.co/RichardErkhov/oopere_-_martra-open-gemma-2b-it-dpo-gguf/blob/main/martra-open-gemma-2b-it-dpo.Q4_K.gguf) | Q4_K | 1.52GB |\n| [martra-open-gemma-2b-it-dpo.Q4_K_M.gguf](https://huggingface.co/RichardErkhov/oopere_-_martra-open-gemma-2b-it-dpo-gguf/blob/main/martra-open-gemma-2b-it-dpo.Q4_K_M.gguf) | Q4_K_M | 1.52GB |\n| [martra-open-gemma-2b-it-dpo.Q4_1.gguf](https://huggingface.co/RichardErkhov/oopere_-_martra-open-gemma-2b-it-dpo-gguf/blob/main/martra-open-gemma-2b-it-dpo.Q4_1.gguf) | Q4_1 | 1.56GB |\n| [martra-open-gemma-2b-it-dpo.Q5_0.gguf](https://huggingface.co/RichardErkhov/oopere_-_martra-open-gemma-2b-it-dpo-gguf/blob/main/martra-open-gemma-2b-it-dpo.Q5_0.gguf) | Q5_0 | 1.68GB |\n| [martra-open-gemma-2b-it-dpo.Q5_K_S.gguf](https://huggingface.co/RichardErkhov/oopere_-_martra-open-gemma-2b-it-dpo-gguf/blob/main/martra-open-gemma-2b-it-dpo.Q5_K_S.gguf) | Q5_K_S | 1.68GB |\n| [martra-open-gemma-2b-it-dpo.Q5_K.gguf](https://huggingface.co/RichardErkhov/oopere_-_martra-open-gemma-2b-it-dpo-gguf/blob/main/martra-open-gemma-2b-it-dpo.Q5_K.gguf) | Q5_K | 1.71GB |\n| [martra-open-gemma-2b-it-dpo.Q5_K_M.gguf](https://huggingface.co/RichardErkhov/oopere_-_martra-open-gemma-2b-it-dpo-gguf/blob/main/martra-open-gemma-2b-it-dpo.Q5_K_M.gguf) | Q5_K_M | 1.71GB |\n| [martra-open-gemma-2b-it-dpo.Q5_1.gguf](https://huggingface.co/RichardErkhov/oopere_-_martra-open-gemma-2b-it-dpo-gguf/blob/main/martra-open-gemma-2b-it-dpo.Q5_1.gguf) | Q5_1 | 1.79GB |\n| [martra-open-gemma-2b-it-dpo.Q6_K.gguf](https://huggingface.co/RichardErkhov/oopere_-_martra-open-gemma-2b-it-dpo-gguf/blob/main/martra-open-gemma-2b-it-dpo.Q6_K.gguf) | Q6_K | 1.92GB |\n| [martra-open-gemma-2b-it-dpo.Q8_0.gguf](https://huggingface.co/RichardErkhov/oopere_-_martra-open-gemma-2b-it-dpo-gguf/blob/main/martra-open-gemma-2b-it-dpo.Q8_0.gguf) | Q8_0 | 2.49GB |\n\n\n\n\nOriginal model description:\n---\nlanguage:\n- en\nlicense: mit\nlibrary_name: transformers\ntags:\n- dpo\n- 2b\n- gemma\ndatasets:\n- argilla/distilabel-capybara-dpo-7k-binarized\npipeline_tag: text-generation\nwidget:\n- text: \"3713841893836/4? \\nLimit your response to mathematical expressions and symbols.\"\n  example_title: 'Return only numbers. '\n- text: A group of 10 people is split into 3 different committees of 3, 4, and 3 people,\n    respectively. In how many ways can this be done?\n  example_title: Solve Problem\n---\n\nYou can see the process with instructions for creating the model in the notebook: [Aligning_DPO_gemma-2b.ipynb](https://github.com/peremartra/Large-Language-Model-Notebooks-Course/blob/main/P2-MHF/Aligning_DPO_open_gemma-2b-it.ipynb)\n\nTo create it, we started with the [google/gemma-2b-it](https://huggingface.co/google/gemma-2b-it) model and applied DPO alignment using the [distilabel-capybara-dpo-7k-binarized dataset](https://huggingface.co/datasets/argilla/distilabel-capybara-dpo-7k-binarized).\n\nGemma is a family of lightweight, state-of-the-art open models from Google, built from the same research and technology used to create the Gemini models. \n\n**Row in the Dataset:**\n\n*{'prompt': '<|user|>\\nAssist me in calculating 9319357631 plus 595. Numbers and symbols only, please.<|end|>\\n<|assistant|>\\n',\n 'chosen': 'The sum of 9319357631 and 595 is 9319358226.<|end|>\\n',\n 'rejected': 'The result of adding 9319357631 and 595 is 9319363626.<|end|>\\n'}*\n\n**Prompt:**\n*3713841893836/4?\nLimit your response to mathematical expressions and symbols.*\n\n**Response from the Base model:**\n\n*To find the result of the division, we can simply divide the given number by 4:\n$$\n\\frac{3713841893836}{4} = 928460473459*\n\n**Response from the fine-tuned model:**\n\n*3713841893836 ÷ 4 = 928460473459*\n\n\nIf you want to see how the model was created, you can check out the [repository](https://github.com/peremartra/Large-Language-Model-Notebooks-Course) where the book's notebooks are kept up-to-date.\n\n",
    "related_quantizations": []
  },
  "tags": [
    "gguf",
    "endpoints_compatible",
    "region:us",
    "conversational"
  ],
  "likes": 0,
  "downloads": 102,
  "gated": false,
  "private": false,
  "last_modified": "2024-08-22T19:02:28.000Z",
  "created_at": "2024-08-22T18:28:18.000Z",
  "pipeline_tag": "",
  "library_name": ""
}

{
  "_id": "66c78342060b6c3335e0c139",
  "id": "RichardErkhov/oopere_-_martra-open-gemma-2b-it-dpo-gguf",
  "modelId": "RichardErkhov/oopere_-_martra-open-gemma-2b-it-dpo-gguf",
  "sha": "c38bdb8238f11d61b1453112947468299d478394",
  "createdAt": "2024-08-22T18:28:18.000Z",
  "lastModified": "2024-08-22T19:02:28.000Z",
  "author": "RichardErkhov",
  "downloads": 102,
  "likes": 0,
  "gated": false,
  "private": false,
  "pipeline_tag": "",
  "library_name": "",
  "siblings_count": 24
}