KoSOLAR-10.7B-mixed-v13

The license is cc-by-nc-4.0.

GAI-LLM/KoSOLAR-10.7B-mixed-v13

Model Details

Model Developers Donghoon Oh, Hanmin Myung, Eunyoung Kim (SK C&C G.AI Eng)

Input Models input text only.

Output Models generate text only.

Model Architecture
GAI-LLM/KoSOLAR-10.7B-mixed-v13 is an auto-regressive language model based on the LLaMA2 transformer architecture.

Base Model yanolja/KoSOLAR-10.7B-v0.1

Training Dataset

• We combined Open Korean Dateset using mixed-strategy.
• We use A100 GPU 80GB * 8, when training.

Model Benchmark

KO-LLM leaderboard

• Follow up as Open KO-LLM LeaderBoard.

Implementation Code

### GAI-LLM/KoSOLAR-10.7B-mixed-v13
from transformers import AutoModelForCausalLM, AutoTokenizer
import torch

repo = "GAI-LLM/KoSOLAR-10.7B-mixed-v13"
model = AutoModelForCausalLM.from_pretrained(
        repo,
        return_dict=True,
        torch_dtype=torch.float16,
        device_map='auto'
)
tokenizer = AutoTokenizer.from_pretrained(repo)