izumi-lab/llama-13b-japanese-lora-v0-1ep

This repo contains a low-rank adapter for LLaMA-13b fit on the llm-japanese-dataset dataset.

You can test this at https://huggingface.co/spaces/izumi-lab/llama-13b-japanese-lora-v0-1ep

This version of the weights was trained with the following hyperparameters:

• Epochs: 1
• Batch size: 130
• Cutoff length: 256
• Learning rate: 3e-4
• Lora r: 4
• Lora target modules: q_proj, v_proj

import torch
from transformers import LlamaForCausalLM, LlamaTokenizer
from peft import PeftModel

base_model = "decapoda-research/llama-13b-hf"
# Please note that the special license of decapoda-research/llama-13b-hf is applied.
model = LlamaForCausalLM.from_pretrained(base_model, torch_dtype=torch.float16)
tokenizer = LlamaTokenizer.from_pretrained(base_model)
model = PeftModel.from_pretrained(
    model,
    "izumi-lab/llama-13b-japanese-lora-v0",
    torch_dtype=torch.float16,
)

To see more latest information, please go to llm.msuzuki.me.

Details

• Japanese Paper: https://jxiv.jst.go.jp/index.php/jxiv/preprint/view/383
• English Paper: https://arxiv.org/abs/2305.12720
• GitHub: https://github.com/masanorihirano/llm-japanese-dataset
• Website: llm.msuzuki.me.

Citation:

@preprint{Hirano2023-llmj,
  title={{llm-japanese-dataset v0: Construction of Japanese Chat Dataset for Large Language Models and its Methodology}},
  autor={Masanori HIRANO and Masahiro SUZUKI and Hiroki SAKAJI},
  doi={10.48550/arXiv.2305.12720},
  archivePrefix={arXiv},
  arxivId={2305.12720},
  year={2023}
}

If you have any inquiries, such as joint research, data provision, various types of support, please email to [email protected] .