Open AlenjandroWang opened 8 months ago
https://github.com/THUDM/SwissArmyTransformer/blob/main/sat/model/official/mixtral_model.py
在这里:
https://github.com/THUDM/SwissArmyTransformer/blob/eb4fac918cc86b304840872d4dccaaaf1b477e37/sat/transformer_defaults.py#L156-L202
https://github.com/THUDM/SwissArmyTransformer/blob/main/sat/model/official/mixtral_model.py