lewdiculous/llama-3-lumimaid-8b-v0.1-oas-gguf-iq-imatrix overview

{
  "metadata": {},
  "card_data": {
    "license": "cc-by-nc-4.0",
    "tags": [
      "roleplay",
      "llama3",
      "sillytavern"
    ],
    "frontmatter": {
      "license": "cc-by-nc-4.0",
      "tags": [
        "roleplay",
        "llama3",
        "sillytavern"
      ]
    },
    "hero_image_url": "https://cdn-uploads.huggingface.co/production/uploads/630dfb008df86f1e5becadc3/d3QMaxy3peFTpSlWdWF-k.png",
    "summary": "> [!IMPORTANT] > **Updated!**  > Version (**v2**) files added! With imatrix data generated from the FP16 and conversions directly from the BF16.  > This is a more disk and compute intensive so lets hope we get GPU inference support for BF16 models in llama.cpp.  > Hopefully avoiding any losses in the model conversion, as has been the recently discussed topic on Llama-3 and GGUF lately.  > If you are able to test them and notice any issues let me know in the discussions. > [!IMPORTANT] > **Relevant:**  > These quants have been done after the fixes from **llama.cpp/pull/6920** have been merged.  > Use **KoboldCpp** version **1.64** or higher, make sure you're up-to-date. > [!TIP] > I apologize for disrupting your experience.  > My upload speeds have been cooked and unstable lately.  > If you **want** and you are **able to**...  > You can **support my various endeavors here (Ko-fi)**.  GGUF-IQ-Imatrix quants for NeverSleep/Llama-3-Lumimaid-8B-v0.1-OAS. **Author:**  \"This model received the Orthogonal Activation Steering treatment, **meaning it will rarely refuse any request.**\" > [!WARNING] > Compatible SillyTavern presets here (simple) or here (Virt's Roleplay Presets - recommended).  > Use the latest version of KoboldCpp. **Use the provided presets for testing.**  > Feedback and support for the Authors is always welcome.  > If there are any issues or questions let me know. > [!NOTE] > For **8GB VRAM** GPUs, I recommend the **Q4_K_M-imat** (4.89 BPW) quant for up to 12288 context sizes. !image/png **Original model information:**",
    "quick_links": [],
    "benchmark_table_html": "",
    "readme_markdown": "---\nlicense: cc-by-nc-4.0\ntags:\n- roleplay\n- llama3\n- sillytavern\n---\n\n> [!IMPORTANT]\n> **Updated!** <br>\n> Version (**v2**) files added! With imatrix data generated from the FP16 and conversions directly from the BF16. <br>\n> This is a more disk and compute intensive so lets hope we get GPU inference support for BF16 models in llama.cpp. <br>\n> Hopefully avoiding any losses in the model conversion, as has been the recently discussed topic on Llama-3 and GGUF lately. <br>\n> If you are able to test them and notice any issues let me know in the discussions.\n\n> [!IMPORTANT]\n> **Relevant:** <br>\n> These quants have been done after the fixes from [**llama.cpp/pull/6920**](https://github.com/ggerganov/llama.cpp/pull/6920) have been merged. <br>\n> Use **KoboldCpp** version **1.64** or higher, make sure you're up-to-date.\n\n> [!TIP]\n> I apologize for disrupting your experience. <br>\n> My upload speeds have been cooked and unstable lately. <br>\n> If you **want** and you are **able to**... <br>\n> You can [**support my various endeavors here (Ko-fi)**](https://ko-fi.com/Lewdiculous). <br>\n\nGGUF-IQ-Imatrix quants for [NeverSleep/Llama-3-Lumimaid-8B-v0.1-OAS](https://huggingface.co/NeverSleep/Llama-3-Lumimaid-8B-v0.1-OAS).\n\n**Author:** <br>\n\"This model received the Orthogonal Activation Steering treatment, **meaning it will rarely refuse any request.**\"\n\n> [!WARNING]\n> Compatible SillyTavern presets [here (simple)](https://huggingface.co/ChaoticNeutrals/Poppy_Porpoise-v0.7-L3-8B/tree/main/Official%20Poppy%20Porpoise%20ST%20Presets) or [here (Virt's Roleplay Presets - recommended)](https://huggingface.co/Virt-io/SillyTavern-Presets). <br>\n> Use the latest version of KoboldCpp. **Use the provided presets for testing.** <br>\n> Feedback and support for the Authors is always welcome. <br>\n> If there are any issues or questions let me know.\n\n> [!NOTE]\n> For **8GB VRAM** GPUs, I recommend the **Q4_K_M-imat** (4.89 BPW) quant for up to 12288 context sizes.\n\n![image/png](https://cdn-uploads.huggingface.co/production/uploads/65d4cf2693a0a3744a27536c/JUxfdTot7v7LTdIGYyzYM.png)\n\n**Original model information:**\n\n## Lumimaid 0.1\n\n<center><div style=\"width: 100%;\">\n    <img src=\"https://cdn-uploads.huggingface.co/production/uploads/630dfb008df86f1e5becadc3/d3QMaxy3peFTpSlWdWF-k.png\" style=\"display: block; margin: auto;\">\n</div></center>\n\nThis model uses the Llama3 **prompting format**\n\nLlama3 trained on our RP datasets, we tried to have a balance between the ERP and the RP, not too horny, but just enough.\n\nWe also added some non-RP dataset, making the model less dumb overall. It should look like a 40%/60% ratio for Non-RP/RP+ERP data.\n\nThis model includes the new Luminae dataset from Ikari.\n\nThis model have received the Orthogonal Activation Steering treatment, meaning it will rarely refuse any request.\n\nIf you consider trying this model please give us some feedback either on the Community tab on hf or on our [Discord Server](https://discord.gg/MtCVRWTZXY).\n\n## Credits:\n- Undi\n- IkariDev\n\n## Description\n\nThis repo contains FP16 files of Lumimaid-8B-v0.1-OAS.\n\nSwitch: [8B](https://huggingface.co/NeverSleep/Llama-3-Lumimaid-8B-v0.1) - [70B](https://huggingface.co/NeverSleep/Llama-3-Lumimaid-70B-v0.1) - [70B-alt](https://huggingface.co/NeverSleep/Llama-3-Lumimaid-70B-v0.1-alt) - [8B-OAS](https://huggingface.co/NeverSleep/Llama-3-Lumimaid-8B-v0.1-OAS) - [70B-OAS](https://huggingface.co/NeverSleep/Llama-3-Lumimaid-70B-v0.1-OAS)\n\n## Training data used:\n- [Aesir datasets](https://huggingface.co/MinervaAI)\n- [NoRobots](https://huggingface.co/datasets/Doctor-Shotgun/no-robots-sharegpt)\n- [limarp](https://huggingface.co/datasets/lemonilia/LimaRP) - 8k ctx\n- [toxic-dpo-v0.1-sharegpt](https://huggingface.co/datasets/Undi95/toxic-dpo-v0.1-sharegpt)\n- [ToxicQAFinal](https://huggingface.co/datasets/NobodyExistsOnTheInternet/ToxicQAFinal)\n- Luminae-i1 (70B/70B-alt) (i2 was not existing when the 70b started training) | Luminae-i2 (8B) (this one gave better results on the 8b) - Ikari's Dataset\n- [Squish42/bluemoon-fandom-1-1-rp-cleaned](https://huggingface.co/datasets/Squish42/bluemoon-fandom-1-1-rp-cleaned) - 50% (randomly)\n- [NobodyExistsOnTheInternet/PIPPAsharegptv2test](https://huggingface.co/datasets/NobodyExistsOnTheInternet/PIPPAsharegptv2test) - 5% (randomly)\n- [cgato/SlimOrcaDedupCleaned](https://huggingface.co/datasets/cgato/SlimOrcaDedupCleaned) - 5% (randomly)\n- Airoboros (reduced)\n- [Capybara](https://huggingface.co/datasets/Undi95/Capybara-ShareGPT/) (reduced)\n\n\n## Models used (only for 8B)\n\n- Initial LumiMaid 8B Finetune\n- Undi95/Llama-3-Unholy-8B-e4\n- Undi95/Llama-3-LewdPlay-8B\n\n## Prompt template: Llama3\n\n```\n<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\n{system_prompt}<|eot_id|><|start_header_id|>user<|end_header_id|>\n\n{input}<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n{output}<|eot_id|>\n```\n\n## Others\n\nUndi: If you want to support us, you can [here](https://ko-fi.com/undiai).\n\nIkariDev: Visit my [retro/neocities style website](https://ikaridevgit.github.io/) please kek",
    "related_quantizations": []
  },
  "tags": [
    "gguf",
    "roleplay",
    "llama3",
    "sillytavern",
    "license:cc-by-nc-4.0",
    "endpoints_compatible",
    "region:us"
  ],
  "likes": 76,
  "downloads": 3779,
  "gated": false,
  "private": false,
  "last_modified": "2024-05-12T11:33:57.000Z",
  "created_at": "2024-05-07T14:15:04.000Z",
  "pipeline_tag": "",
  "library_name": ""
}

{
  "_id": "663a3768d67a4c20c6b83024",
  "id": "Lewdiculous/Llama-3-Lumimaid-8B-v0.1-OAS-GGUF-IQ-Imatrix",
  "modelId": "Lewdiculous/Llama-3-Lumimaid-8B-v0.1-OAS-GGUF-IQ-Imatrix",
  "sha": "a6a65f4d7c1c282134dd5b44f08d05c79f58247a",
  "createdAt": "2024-05-07T14:15:04.000Z",
  "lastModified": "2024-05-12T11:33:57.000Z",
  "author": "Lewdiculous",
  "downloads": 3779,
  "likes": 76,
  "gated": false,
  "private": false,
  "pipeline_tag": "",
  "library_name": "",
  "siblings_count": 28
}