Model Intelligence Sheet
richarderkhov/raincandy-u_-_llama-3-5b-sheard-gguf overview
Pruned version of Llama-3-8b. Tool used: PrunMe, Mergekit. Meta Llama 3 is licensed under the Meta Llama 3 Community License, Copyright © Meta Platforms, Inc. All Rights Reserved.
Downloads
1,921
Likes
0
Pipeline
—
Library
—
Visibility
Public
Access
Open
Repository Files & Downloads
22 files detected
Direct downloads for all repository files
| File | Type | Quantization | Size | Link |
|---|---|---|---|---|
| Llama-3-5B-Sheard.IQ3_M.gguf | GGUF | IQ3_M | 2.61 GB | Download |
| Llama-3-5B-Sheard.IQ3_S.gguf | GGUF | IQ3_S | 2.55 GB | Download |
| Llama-3-5B-Sheard.IQ3_XS.gguf | GGUF | IQ3_XS | 2.44 GB | Download |
| Llama-3-5B-Sheard.IQ4_NL.gguf | GGUF | IQ4_NL | 3.22 GB | Download |
| Llama-3-5B-Sheard.IQ4_XS.gguf | GGUF | IQ4_XS | 3.07 GB | Download |
| Llama-3-5B-Sheard.Q2_K.gguf | GGUF | Q2_K | 2.21 GB | Download |
| Llama-3-5B-Sheard.Q3_K.gguf | GGUF | Q3_K | 2.76 GB | Download |
| Llama-3-5B-Sheard.Q3_K_L.gguf | GGUF | Q3_K_L | 2.96 GB | Download |
| Llama-3-5B-Sheard.Q3_K_M.gguf | GGUF | Q3_K_M | 2.76 GB | Download |
| Llama-3-5B-Sheard.Q3_K_S.gguf | GGUF | Q3_K_S | 2.54 GB | Download |
| Llama-3-5B-Sheard.Q4_0.gguf | GGUF | — | 3.20 GB | Download |
| Llama-3-5B-Sheard.Q4_1.gguf | GGUF | — | 3.51 GB | Download |
| Llama-3-5B-Sheard.Q4_K.gguf | GGUF | Q4_K | 3.35 GB | Download |
| Llama-3-5B-Sheard.Q4_K_M.gguf | GGUF | Q4_K_M | 3.35 GB | Download |
| Llama-3-5B-Sheard.Q4_K_S.gguf | GGUF | Q4_K_S | 3.21 GB | Download |
| Llama-3-5B-Sheard.Q5_0.gguf | GGUF | — | 3.82 GB | Download |
| Llama-3-5B-Sheard.Q5_1.gguf | GGUF | — | 4.13 GB | Download |
| Llama-3-5B-Sheard.Q5_K.gguf | GGUF | Q5_K | 3.90 GB | Download |
| Llama-3-5B-Sheard.Q5_K_M.gguf | GGUF | Q5_K_M | 3.90 GB | Download |
| Llama-3-5B-Sheard.Q5_K_S.gguf | GGUF | Q5_K_S | 3.82 GB | Download |
| Llama-3-5B-Sheard.Q6_K.gguf | GGUF | Q6_K | 4.48 GB | Download |
| Llama-3-5B-Sheard.Q8_0.gguf | GGUF | — | 5.80 GB | Download |
Model Details Live
Metadata Inspector
Normalized metadata (stored in metadata_json)
{
"metadata": {},
"card_data": {
"frontmatter": {},
"hero_image_url": "https://cdn-uploads.huggingface.co/production/uploads/657eb5b256c9c67605a6e8b5/6X2l7zKi9ttuxfRGLucUY.png",
"summary": "Pruned version of Llama-3-8b. Tool used: PrunMe, Mergekit. **Meta Llama 3 is licensed under the Meta Llama 3 Community License, Copyright © Meta Platforms, Inc. All Rights Reserved.**",
"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\nLlama-3-5B-Sheard - GGUF\n- Model creator: https://huggingface.co/raincandy-u/\n- Original model: https://huggingface.co/raincandy-u/Llama-3-5B-Sheard/\n\n\n| Name | Quant method | Size |\n| ---- | ---- | ---- |\n| [Llama-3-5B-Sheard.Q2_K.gguf](https://huggingface.co/RichardErkhov/raincandy-u_-_Llama-3-5B-Sheard-gguf/blob/main/Llama-3-5B-Sheard.Q2_K.gguf) | Q2_K | 2.21GB |\n| [Llama-3-5B-Sheard.IQ3_XS.gguf](https://huggingface.co/RichardErkhov/raincandy-u_-_Llama-3-5B-Sheard-gguf/blob/main/Llama-3-5B-Sheard.IQ3_XS.gguf) | IQ3_XS | 2.44GB |\n| [Llama-3-5B-Sheard.IQ3_S.gguf](https://huggingface.co/RichardErkhov/raincandy-u_-_Llama-3-5B-Sheard-gguf/blob/main/Llama-3-5B-Sheard.IQ3_S.gguf) | IQ3_S | 2.55GB |\n| [Llama-3-5B-Sheard.Q3_K_S.gguf](https://huggingface.co/RichardErkhov/raincandy-u_-_Llama-3-5B-Sheard-gguf/blob/main/Llama-3-5B-Sheard.Q3_K_S.gguf) | Q3_K_S | 2.54GB |\n| [Llama-3-5B-Sheard.IQ3_M.gguf](https://huggingface.co/RichardErkhov/raincandy-u_-_Llama-3-5B-Sheard-gguf/blob/main/Llama-3-5B-Sheard.IQ3_M.gguf) | IQ3_M | 2.61GB |\n| [Llama-3-5B-Sheard.Q3_K.gguf](https://huggingface.co/RichardErkhov/raincandy-u_-_Llama-3-5B-Sheard-gguf/blob/main/Llama-3-5B-Sheard.Q3_K.gguf) | Q3_K | 2.76GB |\n| [Llama-3-5B-Sheard.Q3_K_M.gguf](https://huggingface.co/RichardErkhov/raincandy-u_-_Llama-3-5B-Sheard-gguf/blob/main/Llama-3-5B-Sheard.Q3_K_M.gguf) | Q3_K_M | 2.76GB |\n| [Llama-3-5B-Sheard.Q3_K_L.gguf](https://huggingface.co/RichardErkhov/raincandy-u_-_Llama-3-5B-Sheard-gguf/blob/main/Llama-3-5B-Sheard.Q3_K_L.gguf) | Q3_K_L | 2.96GB |\n| [Llama-3-5B-Sheard.IQ4_XS.gguf](https://huggingface.co/RichardErkhov/raincandy-u_-_Llama-3-5B-Sheard-gguf/blob/main/Llama-3-5B-Sheard.IQ4_XS.gguf) | IQ4_XS | 3.07GB |\n| [Llama-3-5B-Sheard.Q4_0.gguf](https://huggingface.co/RichardErkhov/raincandy-u_-_Llama-3-5B-Sheard-gguf/blob/main/Llama-3-5B-Sheard.Q4_0.gguf) | Q4_0 | 3.2GB |\n| [Llama-3-5B-Sheard.IQ4_NL.gguf](https://huggingface.co/RichardErkhov/raincandy-u_-_Llama-3-5B-Sheard-gguf/blob/main/Llama-3-5B-Sheard.IQ4_NL.gguf) | IQ4_NL | 3.22GB |\n| [Llama-3-5B-Sheard.Q4_K_S.gguf](https://huggingface.co/RichardErkhov/raincandy-u_-_Llama-3-5B-Sheard-gguf/blob/main/Llama-3-5B-Sheard.Q4_K_S.gguf) | Q4_K_S | 3.21GB |\n| [Llama-3-5B-Sheard.Q4_K.gguf](https://huggingface.co/RichardErkhov/raincandy-u_-_Llama-3-5B-Sheard-gguf/blob/main/Llama-3-5B-Sheard.Q4_K.gguf) | Q4_K | 3.35GB |\n| [Llama-3-5B-Sheard.Q4_K_M.gguf](https://huggingface.co/RichardErkhov/raincandy-u_-_Llama-3-5B-Sheard-gguf/blob/main/Llama-3-5B-Sheard.Q4_K_M.gguf) | Q4_K_M | 3.35GB |\n| [Llama-3-5B-Sheard.Q4_1.gguf](https://huggingface.co/RichardErkhov/raincandy-u_-_Llama-3-5B-Sheard-gguf/blob/main/Llama-3-5B-Sheard.Q4_1.gguf) | Q4_1 | 3.51GB |\n| [Llama-3-5B-Sheard.Q5_0.gguf](https://huggingface.co/RichardErkhov/raincandy-u_-_Llama-3-5B-Sheard-gguf/blob/main/Llama-3-5B-Sheard.Q5_0.gguf) | Q5_0 | 3.82GB |\n| [Llama-3-5B-Sheard.Q5_K_S.gguf](https://huggingface.co/RichardErkhov/raincandy-u_-_Llama-3-5B-Sheard-gguf/blob/main/Llama-3-5B-Sheard.Q5_K_S.gguf) | Q5_K_S | 3.82GB |\n| [Llama-3-5B-Sheard.Q5_K.gguf](https://huggingface.co/RichardErkhov/raincandy-u_-_Llama-3-5B-Sheard-gguf/blob/main/Llama-3-5B-Sheard.Q5_K.gguf) | Q5_K | 3.9GB |\n| [Llama-3-5B-Sheard.Q5_K_M.gguf](https://huggingface.co/RichardErkhov/raincandy-u_-_Llama-3-5B-Sheard-gguf/blob/main/Llama-3-5B-Sheard.Q5_K_M.gguf) | Q5_K_M | 3.9GB |\n| [Llama-3-5B-Sheard.Q5_1.gguf](https://huggingface.co/RichardErkhov/raincandy-u_-_Llama-3-5B-Sheard-gguf/blob/main/Llama-3-5B-Sheard.Q5_1.gguf) | Q5_1 | 4.13GB |\n| [Llama-3-5B-Sheard.Q6_K.gguf](https://huggingface.co/RichardErkhov/raincandy-u_-_Llama-3-5B-Sheard-gguf/blob/main/Llama-3-5B-Sheard.Q6_K.gguf) | Q6_K | 4.48GB |\n| [Llama-3-5B-Sheard.Q8_0.gguf](https://huggingface.co/RichardErkhov/raincandy-u_-_Llama-3-5B-Sheard-gguf/blob/main/Llama-3-5B-Sheard.Q8_0.gguf) | Q8_0 | 5.8GB |\n\n\n\n\nOriginal model description:\n---\nlanguage:\n- en\npipeline_tag: text-generation\ntags:\n- facebook\n- meta\n- pytorch\n- llama\n- llama-3\nlicense: other\nlicense_name: llama3\nlicense_link: LICENSE\ndatasets:\n- JeanKaddour/minipile\n- raincandy-u/SlimOrca-Llama-3-Preference-DPO-Pairs\n---\n\n\n\n\n\n# Llama-3-5B-Sheard\n\nPruned version of Llama-3-8b.\n\nTool used: [PrunMe](https://github.com/arcee-ai/PruneMe), Mergekit.\n\n**Meta Llama 3 is licensed under the Meta Llama 3 Community License, Copyright © Meta Platforms, Inc. All Rights Reserved.**\n\n## Training\n\nAfter sliced by mergekit, the model is continue-pretrained on minipile for 1 epoch and ~100k samples. Then we trained it using ORPO on Llama-3-70b generated DPO pairs.\n\n## Disclaimer\n\nThis model is for testing purposes only, and when the system prompt is not empty, the output may repeat and not stop!\n\n<h1><a href=\"https://discord.gg/Dw8SkfBm\">Join our discord</a></h1>\n\n",
"related_quantizations": []
},
"tags": [
"gguf",
"endpoints_compatible",
"region:us",
"conversational"
],
"likes": 0,
"downloads": 1921,
"gated": false,
"private": false,
"last_modified": "2024-06-30T04:14:46.000Z",
"created_at": "2024-06-30T02:06:32.000Z",
"pipeline_tag": "",
"library_name": ""
}
Source payload excerpt (from Hugging Face API)
{
"_id": "6680bda8f911bc2f797c39b0",
"id": "RichardErkhov/raincandy-u_-_Llama-3-5B-Sheard-gguf",
"modelId": "RichardErkhov/raincandy-u_-_Llama-3-5B-Sheard-gguf",
"sha": "cb5238e2b4a31ae63549f0fe3e8a55de2702f976",
"createdAt": "2024-06-30T02:06:32.000Z",
"lastModified": "2024-06-30T04:14:46.000Z",
"author": "RichardErkhov",
"downloads": 1921,
"likes": 0,
"gated": false,
"private": false,
"pipeline_tag": "",
"library_name": "",
"siblings_count": 24
}