valueeval24-bert-baseline-en

This model is a fine-tuned version of bert-base-uncased on the ValueEval'24 dataset. It achieves the following results on the evaluation set:

• Loss: 0.0631
• F1-score: {'Self-direction: thought attained': 0, 'Self-direction: thought constrained': 0, 'Self-direction: action attained': 0.1, 'Self-direction: action constrained': 0, 'Stimulation attained': 0.27, 'Stimulation constrained': 0, 'Hedonism attained': 0.27, 'Hedonism constrained': 0, 'Achievement attained': 0.32, 'Achievement constrained': 0.17, 'Power: dominance attained': 0.3, 'Power: dominance constrained': 0.03, 'Power: resources attained': 0.23, 'Power: resources constrained': 0.22, 'Face attained': 0, 'Face constrained': 0.06, 'Security: personal attained': 0.04, 'Security: personal constrained': 0.25, 'Security: societal attained': 0.3, 'Security: societal constrained': 0.4, 'Tradition attained': 0.41, 'Tradition constrained': 0, 'Conformity: rules attained': 0.37, 'Conformity: rules constrained': 0.29, 'Conformity: interpersonal attained': 0, 'Conformity: interpersonal constrained': 0, 'Humility attained': 0, 'Humility constrained': 0, 'Benevolence: caring attained': 0.14, 'Benevolence: caring constrained': 0, 'Benevolence: dependability attained': 0.16, 'Benevolence: dependability constrained': 0, 'Universalism: concern attained': 0.32, 'Universalism: concern constrained': 0.12, 'Universalism: nature attained': 0.47, 'Universalism: nature constrained': 0.35, 'Universalism: tolerance attained': 0, 'Universalism: tolerance constrained': 0}
• Marco-avg-f1-score: 0.15

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: 8
• eval_batch_size: 8
• 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: linear
• num_epochs: 5

Training results

Training Loss	Epoch	Step	Validation Loss	F1-score	Marco-avg-f1-score
0.0667	1.0	5595	0.0621	{'Self-direction: thought attained': 0, 'Self-direction: thought constrained': 0, 'Self-direction: action attained': 0, 'Self-direction: action constrained': 0, 'Stimulation attained': 0, 'Stimulation constrained': 0, 'Hedonism attained': 0, 'Hedonism constrained': 0, 'Achievement attained': 0.01, 'Achievement constrained': 0, 'Power: dominance attained': 0.01, 'Power: dominance constrained': 0, 'Power: resources attained': 0, 'Power: resources constrained': 0, 'Face attained': 0, 'Face constrained': 0, 'Security: personal attained': 0, 'Security: personal constrained': 0, 'Security: societal attained': 0.13, 'Security: societal constrained': 0.18, 'Tradition attained': 0, 'Tradition constrained': 0, 'Conformity: rules attained': 0.02, 'Conformity: rules constrained': 0, 'Conformity: interpersonal attained': 0, 'Conformity: interpersonal constrained': 0, 'Humility attained': 0, 'Humility constrained': 0, 'Benevolence: caring attained': 0, 'Benevolence: caring constrained': 0, 'Benevolence: dependability attained': 0, 'Benevolence: dependability constrained': 0, 'Universalism: concern attained': 0, 'Universalism: concern constrained': 0, 'Universalism: nature attained': 0, 'Universalism: nature constrained': 0, 'Universalism: tolerance attained': 0, 'Universalism: tolerance constrained': 0}	0.01
0.0564	2.0	11190	0.0586	{'Self-direction: thought attained': 0, 'Self-direction: thought constrained': 0, 'Self-direction: action attained': 0, 'Self-direction: action constrained': 0, 'Stimulation attained': 0.09, 'Stimulation constrained': 0, 'Hedonism attained': 0, 'Hedonism constrained': 0, 'Achievement attained': 0.23, 'Achievement constrained': 0.05, 'Power: dominance attained': 0.16, 'Power: dominance constrained': 0, 'Power: resources attained': 0.12, 'Power: resources constrained': 0.02, 'Face attained': 0, 'Face constrained': 0, 'Security: personal attained': 0, 'Security: personal constrained': 0.02, 'Security: societal attained': 0.3, 'Security: societal constrained': 0.4, 'Tradition attained': 0.03, 'Tradition constrained': 0, 'Conformity: rules attained': 0.29, 'Conformity: rules constrained': 0.15, 'Conformity: interpersonal attained': 0, 'Conformity: interpersonal constrained': 0, 'Humility attained': 0, 'Humility constrained': 0, 'Benevolence: caring attained': 0, 'Benevolence: caring constrained': 0, 'Benevolence: dependability attained': 0.01, 'Benevolence: dependability constrained': 0, 'Universalism: concern attained': 0.29, 'Universalism: concern constrained': 0.01, 'Universalism: nature attained': 0.46, 'Universalism: nature constrained': 0, 'Universalism: tolerance attained': 0, 'Universalism: tolerance constrained': 0}	0.07
0.0482	3.0	16785	0.0607	{'Self-direction: thought attained': 0, 'Self-direction: thought constrained': 0, 'Self-direction: action attained': 0.06, 'Self-direction: action constrained': 0, 'Stimulation attained': 0.23, 'Stimulation constrained': 0, 'Hedonism attained': 0.27, 'Hedonism constrained': 0, 'Achievement attained': 0.34, 'Achievement constrained': 0.21, 'Power: dominance attained': 0.3, 'Power: dominance constrained': 0.01, 'Power: resources attained': 0.2, 'Power: resources constrained': 0.25, 'Face attained': 0, 'Face constrained': 0.02, 'Security: personal attained': 0, 'Security: personal constrained': 0.26, 'Security: societal attained': 0.25, 'Security: societal constrained': 0.4, 'Tradition attained': 0.43, 'Tradition constrained': 0, 'Conformity: rules attained': 0.34, 'Conformity: rules constrained': 0.26, 'Conformity: interpersonal attained': 0, 'Conformity: interpersonal constrained': 0, 'Humility attained': 0, 'Humility constrained': 0, 'Benevolence: caring attained': 0.08, 'Benevolence: caring constrained': 0, 'Benevolence: dependability attained': 0.14, 'Benevolence: dependability constrained': 0, 'Universalism: concern attained': 0.29, 'Universalism: concern constrained': 0.11, 'Universalism: nature attained': 0.48, 'Universalism: nature constrained': 0.37, 'Universalism: tolerance attained': 0, 'Universalism: tolerance constrained': 0}	0.14
0.0393	4.0	22380	0.0631	{'Self-direction: thought attained': 0, 'Self-direction: thought constrained': 0, 'Self-direction: action attained': 0.1, 'Self-direction: action constrained': 0, 'Stimulation attained': 0.27, 'Stimulation constrained': 0, 'Hedonism attained': 0.27, 'Hedonism constrained': 0, 'Achievement attained': 0.32, 'Achievement constrained': 0.17, 'Power: dominance attained': 0.3, 'Power: dominance constrained': 0.03, 'Power: resources attained': 0.23, 'Power: resources constrained': 0.22, 'Face attained': 0, 'Face constrained': 0.06, 'Security: personal attained': 0.04, 'Security: personal constrained': 0.25, 'Security: societal attained': 0.3, 'Security: societal constrained': 0.4, 'Tradition attained': 0.41, 'Tradition constrained': 0, 'Conformity: rules attained': 0.37, 'Conformity: rules constrained': 0.29, 'Conformity: interpersonal attained': 0, 'Conformity: interpersonal constrained': 0, 'Humility attained': 0, 'Humility constrained': 0, 'Benevolence: caring attained': 0.14, 'Benevolence: caring constrained': 0, 'Benevolence: dependability attained': 0.16, 'Benevolence: dependability constrained': 0, 'Universalism: concern attained': 0.32, 'Universalism: concern constrained': 0.12, 'Universalism: nature attained': 0.47, 'Universalism: nature constrained': 0.35, 'Universalism: tolerance attained': 0, 'Universalism: tolerance constrained': 0}	0.15
0.0337	5.0	27975	0.0652	{'Self-direction: thought attained': 0, 'Self-direction: thought constrained': 0, 'Self-direction: action attained': 0.14, 'Self-direction: action constrained': 0, 'Stimulation attained': 0.25, 'Stimulation constrained': 0, 'Hedonism attained': 0.27, 'Hedonism constrained': 0, 'Achievement attained': 0.33, 'Achievement constrained': 0.2, 'Power: dominance attained': 0.3, 'Power: dominance constrained': 0.03, 'Power: resources attained': 0.24, 'Power: resources constrained': 0.23, 'Face attained': 0.01, 'Face constrained': 0.11, 'Security: personal attained': 0.05, 'Security: personal constrained': 0.23, 'Security: societal attained': 0.3, 'Security: societal constrained': 0.41, 'Tradition attained': 0.38, 'Tradition constrained': 0, 'Conformity: rules attained': 0.36, 'Conformity: rules constrained': 0.29, 'Conformity: interpersonal attained': 0, 'Conformity: interpersonal constrained': 0, 'Humility attained': 0, 'Humility constrained': 0, 'Benevolence: caring attained': 0.18, 'Benevolence: caring constrained': 0, 'Benevolence: dependability attained': 0.17, 'Benevolence: dependability constrained': 0, 'Universalism: concern attained': 0.33, 'Universalism: concern constrained': 0.12, 'Universalism: nature attained': 0.46, 'Universalism: nature constrained': 0.35, 'Universalism: tolerance attained': 0, 'Universalism: tolerance constrained': 0}	0.15

Framework versions

• Transformers 4.53.0
• Pytorch 2.5.1+cu121
• Datasets 3.6.0
• Tokenizers 0.21.2