README.md

metadata

language:
  - en
pipeline_tag: text-generation
license: cc-by-nc-sa-4.0
datasets:
  - databricks/databricks-dolly-15k
  - kyujinpy/KOR-OpenOrca-Platypus-v3

Input Models input text only.

Output Models generate text only.

Base Model beomi/Yi-Ko-6B

Implementation Code

from transformers import AutoModelForCausalLM, AutoTokenizer
import torch
repo = "ifuseok/yi-ko-playtus-instruct-v0.2"
OpenOrca = AutoModelForCausalLM.from_pretrained(
        repo,
        return_dict=True,
        torch_dtype=torch.float16,
        device_map='auto'
)
OpenOrca_tokenizer = AutoTokenizer.from_pretrained(repo)

Prompt Example

<|system|> 시스템 메시지 입니다. <|endoftext|> <|user|> 유저 입니다.<|endoftext|> <|assistant|> 어시스턴트 입니다.<|endoftext|>