Quantization made by Richard Erkhov.

Llama-3-neoAI-8B-Chat-v0.1 - GGUF

Model creator: https://huggingface.co/neoai-inc/
Original model: https://huggingface.co/neoai-inc/Llama-3-neoAI-8B-Chat-v0.1/

Name	Quant method	Size
Llama-3-neoAI-8B-Chat-v0.1.Q2_K.gguf	Q2_K	2.96GB
Llama-3-neoAI-8B-Chat-v0.1.IQ3_XS.gguf	IQ3_XS	3.28GB
Llama-3-neoAI-8B-Chat-v0.1.IQ3_S.gguf	IQ3_S	3.43GB
Llama-3-neoAI-8B-Chat-v0.1.Q3_K_S.gguf	Q3_K_S	3.41GB
Llama-3-neoAI-8B-Chat-v0.1.IQ3_M.gguf	IQ3_M	3.52GB
Llama-3-neoAI-8B-Chat-v0.1.Q3_K.gguf	Q3_K	3.74GB
Llama-3-neoAI-8B-Chat-v0.1.Q3_K_M.gguf	Q3_K_M	3.74GB
Llama-3-neoAI-8B-Chat-v0.1.Q3_K_L.gguf	Q3_K_L	4.03GB
Llama-3-neoAI-8B-Chat-v0.1.IQ4_XS.gguf	IQ4_XS	4.18GB
Llama-3-neoAI-8B-Chat-v0.1.Q4_0.gguf	Q4_0	4.34GB
Llama-3-neoAI-8B-Chat-v0.1.IQ4_NL.gguf	IQ4_NL	4.38GB
Llama-3-neoAI-8B-Chat-v0.1.Q4_K_S.gguf	Q4_K_S	4.37GB
Llama-3-neoAI-8B-Chat-v0.1.Q4_K.gguf	Q4_K	4.58GB
Llama-3-neoAI-8B-Chat-v0.1.Q4_K_M.gguf	Q4_K_M	4.58GB
Llama-3-neoAI-8B-Chat-v0.1.Q4_1.gguf	Q4_1	4.78GB
Llama-3-neoAI-8B-Chat-v0.1.Q5_0.gguf	Q5_0	5.21GB
Llama-3-neoAI-8B-Chat-v0.1.Q5_K_S.gguf	Q5_K_S	5.21GB
Llama-3-neoAI-8B-Chat-v0.1.Q5_K.gguf	Q5_K	5.34GB
Llama-3-neoAI-8B-Chat-v0.1.Q5_K_M.gguf	Q5_K_M	5.34GB
Llama-3-neoAI-8B-Chat-v0.1.Q5_1.gguf	Q5_1	5.65GB
Llama-3-neoAI-8B-Chat-v0.1.Q6_K.gguf	Q6_K	6.14GB
Llama-3-neoAI-8B-Chat-v0.1.Q8_0.gguf	Q8_0	7.95GB

Original model description:

license: llama3 language: - ja - en pipeline_tag: text-generation

Llama 3 neoAI 8B Chat v0.1

Model Description

Llama 3 neoAI 8B Chat v0.1は，Meta-Llama-3-8B-Instructをベースとして日本語能力を強化するために事後学習を行なったモデルです．

詳細はブログ記事を参照してください．

Evaluation

Usage

Huggingface

import transformers
import torch

model_id = "neoai-inc/Llama-3-neoAI-8B-Chat-v0.1"

pipeline = transformers.pipeline(
    "text-generation",
    model=model_id,
    model_kwargs={"torch_dtype": torch.bfloat16},
    device_map="auto",
)

messages = [
    {"role": "system", "content": "あなたは誠実で優秀な日本人のアシスタントです。"},
    {"role": "user", "content": "日本で一番高い山は？"},
]

terminators = [
    pipeline.tokenizer.eos_token_id,
    pipeline.tokenizer.convert_tokens_to_ids("<|eot_id|>")
]

outputs = pipeline(
    messages,
    max_new_tokens=256,
    eos_token_id=terminators,
    do_sample=True,
    temperature=0.6,
    top_p=0.9,
)
print(outputs[0]["generated_text"][-1])

License

Developed by

Developers

以下アルファベット順

How to Cite

@misc{neoAI-Llama3-8B-Chat-v0.1, 
      title={Llama-3-neoAI-8B-Chat-v0.1}, 
      url={https://huggingface.co/neoai-inc/Llama-3-neoAI-8B-Chat-v0.1}, 
      author={Gouki Minegishi and Koki Itai and Koshiro Terasawa and Masaki Otsuki and Ryo Yagi},
      year={2024},
}