--- base_model: - tokyotech-llm/Llama-3.1-Swallow-8B-Instruct-v0.1 - NousResearch/Hermes-3-Llama-3.1-8B tags: - merge - mergekit language: - ja - en library_name: transformers --- # Llama 3.1 Swallow Hermes 8B v0.1 Llama 3.1 Swallow Hermes 8B v0.1 is a merge of the following models using [Mergekit](https://github.com/arcee-ai/mergekit): * [tokyotech-llm/Llama-3.1-Swallow-8B-Instruct-v0.1](https://huggingface.co/tokyotech-llm/Llama-3.1-Swallow-8B-Instruct-v0.1) * [NousResearch/Hermes-3-Llama-3.1-8B](https://huggingface.co/NousResearch/Hermes-3-Llama-3.1-8B) ## 🧩 Configuration ```yaml base_model: tokyotech-llm/Llama-3.1-Swallow-8B-Instruct-v0.1 models: - model: tokyotech-llm/Llama-3.1-Swallow-8B-Instruct-v0.1 # No parameters necessary for base model - model: NousResearch/Hermes-3-Llama-3.1-8B parameters: density: 0.525 weight: 1 merge_method: dare_ties tokenizer_source: union dtype: bfloat16 ``` ## 💻 Usage ```python !pip install -qU transformers accelerate from transformers import AutoTokenizer import transformers import torch model = "AELLM/llama-3.1-swallow-hermes-8b-v0.1" messages = [{"role": "user", "content": "あなたはタイムトラベラーのAIアシスタントだ。タイムパラドックスを引き起こすことなく、異なる歴史的時代に溶け込むにはどうすればいいのか?"}] tokenizer = AutoTokenizer.from_pretrained(model) prompt = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True) pipeline = transformers.pipeline( "text-generation", model=model, torch_dtype=torch.float16, device_map="auto", ) outputs = pipeline(prompt, max_new_tokens=256, do_sample=True, temperature=0.7, top_k=50, top_p=0.95) print(outputs[0]["generated_text"]) ```