legal_mt5_small_finetuned

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: 0.2780

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: 3e-05
• train_batch_size: 4
• eval_batch_size: 4
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 16
• optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 500
• num_epochs: 10

Training results

Training Loss	Epoch	Step	Validation Loss
5.226	1.1039	500	2.0588
1.4832	2.2077	1000	0.7003
0.787	3.3116	1500	0.4511
0.6142	4.4155	2000	0.3904
0.4991	5.5193	2500	0.3456
0.4446	6.6232	3000	0.3153
0.4441	7.7271	3500	0.2896
0.3978	8.8309	4000	0.2823
0.3954	9.9348	4500	0.2780

Framework versions

• Transformers 4.48.2
• Pytorch 2.5.1+cu124
• Datasets 3.2.0
• Tokenizers 0.21.0