t5-fine-tuned-embedding-chunking

This model is a fine-tuned version of t5-base on the None dataset. It achieves the following results on the evaluation set:

• Loss: 1.6113
• Rouge1: 0.2152
• Rouge2: 0.102
• Rougel: 0.193
• Rougelsum: 0.193
• Bertscore Precision: 0.8727
• Bertscore Recall: 0.8246
• Bertscore F1: 0.8477

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: 2
• eval_batch_size: 2
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 8
• optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• num_epochs: 6

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum	Bertscore Precision	Bertscore Recall	Bertscore F1
2.0877	1.0	914	1.5922	0.2495	0.0981	0.2094	0.2093	0.8753	0.8198	0.8465
1.2277	2.0	1828	1.5686	0.2087	0.0885	0.1844	0.1843	0.8676	0.8213	0.8435
1.0059	3.0	2742	1.5758	0.2277	0.1021	0.1993	0.1992	0.8746	0.8239	0.8482
0.9106	4.0	3656	1.5946	0.2118	0.0977	0.1876	0.1875	0.8718	0.8244	0.8471
0.8652	5.0	4570	1.5985	0.217	0.1013	0.1937	0.1936	0.8715	0.8235	0.8465
0.8386	6.0	5484	1.6113	0.2152	0.102	0.193	0.193	0.8727	0.8246	0.8477

Framework versions

• Transformers 4.47.0
• Pytorch 2.5.1+cu121
• Datasets 3.2.0
• Tokenizers 0.21.0