This model is a fine-tuned version of Einmalumdiewelt/DistilBART_CNN_GNAD_V3 on an unknown dataset.
It achieves the following results on the evaluation set:
Loss: 2.8786
Rouge1: 27.0297
Rouge2: 8.1224
Rougel: 17.9777
Rougelsum: 22.8827
Gen Len: 90.3667
Model description
More information needed
Intended uses & limitations
More information needed
Training and evaluation data
More information needed
Training procedure
Training hyperparameters
The following hyperparameters were used during training:
learning_rate: 5e-05
train_batch_size: 8
eval_batch_size: 8
seed: 42
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
