amsterdam_100bm25_passages

This model is a fine-tuned version of google/mt5-small on the None dataset. It achieves the following results on the evaluation set:

• Loss: 5.5628
• Rouge1: 0.1219
• Rouge2: 0.0255
• Rougel: 0.1106
• Rougelsum: 0.1112
• Gen Len: 19.0

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: 2e-05
• train_batch_size: 1
• eval_batch_size: 1
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum	Gen Len
No log	1.0	58	19.3812	0.0	0.0	0.0	0.0	4.0
No log	2.0	116	16.2740	0.0053	0.0	0.0053	0.0053	4.4667
No log	3.0	174	14.9688	0.0197	0.0042	0.0198	0.0199	5.6667
No log	4.0	232	11.6524	0.02	0.0073	0.0177	0.018	6.6667
No log	5.0	290	9.4875	0.0255	0.0073	0.0226	0.0231	7.8667
No log	6.0	348	8.0678	0.0503	0.0136	0.0493	0.049	10.6
No log	7.0	406	6.4898	0.0976	0.0167	0.0858	0.0866	14.9333
No log	8.0	464	6.0317	0.113	0.0232	0.0995	0.1001	18.9333
13.519	9.0	522	5.6579	0.117	0.0238	0.1037	0.1046	19.0
13.519	10.0	580	5.5628	0.1219	0.0255	0.1106	0.1112	19.0

Framework versions

• Transformers 4.28.0
• Pytorch 2.0.1+cu118
• Datasets 2.12.0
• Tokenizers 0.13.3