usakha/Bart_GovReport_model
025
Hyperparameters
learningrate=2e-5 perdevicetrainbatchsize=14 perdeviceevalbatchsize=14 weightdecay=0.01 savetotallimit=3 numtrainepochs=3 predictwithgenerate=True fp16=True
Training Output
globalstep=3003, trainingloss=1.8524150695953217, metrics={'trainruntime': 2319.7329, 'trainsamplespersecond': 18.122, 'trainstepspersecond': 1.295, 'totalflos': 9.110291036818637e+16, 'train_loss': 1.8524150695953217, 'epoch': 3.0}