An Integrated Framework for LLM Pre-training, Continued Pre-training, Supervised Fine-tuning, Reinforcement Learning, and Evaluation
Note: You might need to use different virtual environments for different modules of the project.
-
Pre-training / Continued Pre-training: Details in
./pretrain -
Supervised Fine-tuning: Details in
./finetune -
Reinforcement Learning: Details in
./rl -
Evaluation: Details in
./evaluation