duyntnet/llama-3-8b-synthia-v3.5-imatrix-gguf overview

{
  "metadata": {},
  "card_data": {
    "license": "other",
    "language": [
      "en"
    ],
    "pipeline_tag": "text-generation",
    "inference": false,
    "tags": [
      "transformers",
      "gguf",
      "imatrix",
      "Llama-3-8B-Synthia-v3.5"
    ],
    "frontmatter": {
      "license": "other",
      "language": [
        "en"
      ],
      "pipeline_tag": "text-generation",
      "inference": "false",
      "tags": [
        "transformers",
        "gguf",
        "imatrix",
        "Llama-3-8B-Synthia-v3.5"
      ]
    },
    "hero_image_url": "",
    "summary": "",
    "quick_links": [],
    "benchmark_table_html": "",
    "readme_markdown": "---\nlicense: other\nlanguage:\n- en\npipeline_tag: text-generation\ninference: false\ntags:\n- transformers\n- gguf\n- imatrix\n- Llama-3-8B-Synthia-v3.5\n---\nQuantizations of https://huggingface.co/migtissera/Llama-3-8B-Synthia-v3.5\n\n\n# From original readme\n\n## Sample code to run inference\n\n```python\nimport torch, json\nfrom transformers import AutoModelForCausalLM, AutoTokenizer\n\nmodel_path = \"/home/migel/Tess-2.0-Llama-3-8B\"\noutput_file_path = \"/home/migel/conversations.jsonl\"\n\nmodel = AutoModelForCausalLM.from_pretrained(\n    model_path,\n    torch_dtype=torch.float16,\n    device_map=\"auto\",\n    load_in_4bit=False,\n    trust_remote_code=False,\n)\n\ntokenizer = AutoTokenizer.from_pretrained(model_path, trust_remote_code=True)\n\ndef generate_text(instruction):\n    tokens = tokenizer.encode(instruction)\n    tokens = torch.LongTensor(tokens).unsqueeze(0)\n    tokens = tokens.to(\"cuda\")\n\n    instance = {\n        \"input_ids\": tokens,\n        \"top_p\": 1.0,\n        \"temperature\": 0.75,\n        \"generate_len\": 1024,\n        \"top_k\": 50,\n    }\n\n    length = len(tokens[0])\n    with torch.no_grad():\n        rest = model.generate(\n            input_ids=tokens,\n            max_length=length + instance[\"generate_len\"],\n            use_cache=True,\n            do_sample=True,\n            top_p=instance[\"top_p\"],\n            temperature=instance[\"temperature\"],\n            top_k=instance[\"top_k\"],\n            num_return_sequences=1,\n            pad_token_id=tokenizer.eos_token_id,\n        )\n    output = rest[0][length:]\n    string = tokenizer.decode(output, skip_special_tokens=True)\n    return f\"{string}\"\n\nconversation = \"\"\"<|begin_of_text|><|start_header_id|>system<|end_header_id|>\\n\\nYou are Synthia, a helful, female AI assitant. You always provide detailed answers without hesitation.<|eot_id|><|start_header_id|>user<|end_header_id|>\\n\\n\"\"\"\n\nwhile True:\n    user_input = input(\"You: \")\n    llm_prompt = f\"{conversation}{user_input}<|eot_id|><|start_header_id|>assistant<|end_header_id|>\\n\\n\"\n    answer = generate_text(llm_prompt)\n    print(answer)\n\n    conversation = f\"{llm_prompt}{answer}<|eot_id|><|start_header_id|>user<|end_header_id|>\\n\\n\"\n\n    json_data = {\"prompt\": user_input, \"answer\": answer}\n\n    with open(output_file_path, \"a\") as output_file:\n        output_file.write(json.dumps(json_data) + \"\\n\")\n```",
    "related_quantizations": []
  },
  "tags": [
    "transformers",
    "gguf",
    "imatrix",
    "Llama-3-8B-Synthia-v3.5",
    "text-generation",
    "en",
    "license:other",
    "region:us",
    "conversational"
  ],
  "likes": 0,
  "downloads": 83,
  "gated": false,
  "private": false,
  "last_modified": "2024-06-06T02:38:19.000Z",
  "created_at": "2024-06-05T22:46:49.000Z",
  "pipeline_tag": "text-generation",
  "library_name": "transformers"
}

{
  "_id": "6660ead9f57dbfbdc33ce476",
  "id": "duyntnet/Llama-3-8B-Synthia-v3.5-imatrix-GGUF",
  "modelId": "duyntnet/Llama-3-8B-Synthia-v3.5-imatrix-GGUF",
  "sha": "2885ddc86b5ca4b63e4075bbb7e8f270cdda5c7f",
  "createdAt": "2024-06-05T22:46:49.000Z",
  "lastModified": "2024-06-06T02:38:19.000Z",
  "author": "duyntnet",
  "downloads": 83,
  "likes": 0,
  "gated": false,
  "private": false,
  "pipeline_tag": "text-generation",
  "library_name": "transformers",
  "siblings_count": 29
}