Longformer-es-mental-large_v2

This model is a fine-tuned version of ELiRF/Longformer-es-mental-large on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 1.7790
• Model Preparation Time: 0.0062

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: 1
• eval_batch_size: 1
• seed: 42
• distributed_type: multi-GPU
• num_devices: 4
• gradient_accumulation_steps: 8
• total_train_batch_size: 32
• total_eval_batch_size: 4
• 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
• lr_scheduler_warmup_ratio: 0.06
• num_epochs: 3.0
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Model Preparation Time
3.6303	0.5481	1000	3.0334	0.0062
2.6087	1.0959	2000	2.1358	0.0062
2.2486	1.6440	3000	1.9304	0.0062
2.0714	2.1918	4000	1.8290	0.0062
1.9916	2.7399	5000	1.7794	0.0062

Framework versions

• Transformers 4.57.6
• Pytorch 2.6.0+cu118
• Datasets 4.8.5
• Tokenizers 0.22.2