andreagasparini/ModernBERT-base-stress

results

This model is a fine-tuned version of answerdotai/ModernBERT-base on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.6851
• F1: 0.8263
• F1 Macro: 0.8254
• F1 Micro: 0.8275
• Precision: 0.8312
• Precision Macro: 0.8328
• Precision Micro: 0.8275
• Recall: 0.8275
• Recall Macro: 0.8241
• Recall Micro: 0.8275

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: 16
• eval_batch_size: 16
• seed: 42
• optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: constant
• num_epochs: 3

Training results

Training Loss	Epoch	Step	Validation Loss	F1	F1 Macro	F1 Micro	Precision	Precision Macro	Precision Micro	Recall	Recall Macro	Recall Micro
No log	1.0	160	0.3593	0.8310	0.8307	0.8310	0.8312	0.8305	0.8310	0.8310	0.8309	0.8310
No log	2.0	320	0.4292	0.8290	0.8279	0.8310	0.8394	0.8419	0.8310	0.8310	0.8264	0.8310
No log	3.0	480	0.6851	0.8263	0.8254	0.8275	0.8312	0.8328	0.8275	0.8275	0.8241	0.8275

Framework versions

• Transformers 4.48.0
• Pytorch 2.5.0+cu124
• Datasets 2.16.1
• Tokenizers 0.21.1