├── COMMERCIAL_LICENSE ├── COMMUNITY_LICENSE ├── LICENSE ├── README.md ├── README_EN.md ├── REGISTRATION_INFORMATION ├── REGISTRATION_INFORMATION_EN ├── assets ├── compression_rate.png ├── data_distribution.jpg ├── data_process.png ├── language_distribution.jpg ├── loss.png ├── yayi_dark.png ├── yayi_dark_small.png └── yayi_light.png ├── config ├── deepspeed.json └── hostfile ├── data └── yayi_train_example.json ├── requirements.txt ├── scripts ├── start.sh └── start_lora.sh └── training ├── trainer_chatml.py └── trainer_yayi2.py /COMMERCIAL_LICENSE: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wenge-research/YAYI2/HEAD/COMMERCIAL_LICENSE -------------------------------------------------------------------------------- /COMMUNITY_LICENSE: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wenge-research/YAYI2/HEAD/COMMUNITY_LICENSE -------------------------------------------------------------------------------- /LICENSE: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wenge-research/YAYI2/HEAD/LICENSE -------------------------------------------------------------------------------- /README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wenge-research/YAYI2/HEAD/README.md -------------------------------------------------------------------------------- /README_EN.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wenge-research/YAYI2/HEAD/README_EN.md -------------------------------------------------------------------------------- /REGISTRATION_INFORMATION: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wenge-research/YAYI2/HEAD/REGISTRATION_INFORMATION -------------------------------------------------------------------------------- /REGISTRATION_INFORMATION_EN: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wenge-research/YAYI2/HEAD/REGISTRATION_INFORMATION_EN -------------------------------------------------------------------------------- /assets/compression_rate.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wenge-research/YAYI2/HEAD/assets/compression_rate.png -------------------------------------------------------------------------------- /assets/data_distribution.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wenge-research/YAYI2/HEAD/assets/data_distribution.jpg -------------------------------------------------------------------------------- /assets/data_process.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wenge-research/YAYI2/HEAD/assets/data_process.png -------------------------------------------------------------------------------- /assets/language_distribution.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wenge-research/YAYI2/HEAD/assets/language_distribution.jpg -------------------------------------------------------------------------------- /assets/loss.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wenge-research/YAYI2/HEAD/assets/loss.png -------------------------------------------------------------------------------- /assets/yayi_dark.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wenge-research/YAYI2/HEAD/assets/yayi_dark.png -------------------------------------------------------------------------------- /assets/yayi_dark_small.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wenge-research/YAYI2/HEAD/assets/yayi_dark_small.png -------------------------------------------------------------------------------- /assets/yayi_light.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wenge-research/YAYI2/HEAD/assets/yayi_light.png -------------------------------------------------------------------------------- /config/deepspeed.json: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wenge-research/YAYI2/HEAD/config/deepspeed.json -------------------------------------------------------------------------------- /config/hostfile: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wenge-research/YAYI2/HEAD/config/hostfile -------------------------------------------------------------------------------- /data/yayi_train_example.json: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wenge-research/YAYI2/HEAD/data/yayi_train_example.json -------------------------------------------------------------------------------- /requirements.txt: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wenge-research/YAYI2/HEAD/requirements.txt -------------------------------------------------------------------------------- /scripts/start.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wenge-research/YAYI2/HEAD/scripts/start.sh -------------------------------------------------------------------------------- /scripts/start_lora.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wenge-research/YAYI2/HEAD/scripts/start_lora.sh -------------------------------------------------------------------------------- /training/trainer_chatml.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wenge-research/YAYI2/HEAD/training/trainer_chatml.py -------------------------------------------------------------------------------- /training/trainer_yayi2.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wenge-research/YAYI2/HEAD/training/trainer_yayi2.py --------------------------------------------------------------------------------