JohannesKiesel/valueeval24-bert-baseline-en-2024-02-15

<!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. -->

valueeval24-bert-baseline-en-2024-02-15

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

• —Loss: 0.0458
• —F1-score: {'Self-direction: thought attained': 0, 'Self-direction: thought constrained': 0, 'Self-direction: action attained': 0.16, 'Self-direction: action constrained': 0.24, 'Stimulation attained': 0, 'Stimulation constrained': 0, 'Hedonism attained': 0, 'Hedonism constrained': 0, 'Achievement attained': 0.42, 'Achievement constrained': 0.31, 'Power: dominance attained': 0.21, 'Power: dominance constrained': 0, 'Power: resources attained': 0.2, 'Power: resources constrained': 0, 'Face attained': 0, 'Face constrained': 0.33, 'Security: personal attained': 0, 'Security: personal constrained': 0.06, 'Security: societal attained': 0.36, 'Security: societal constrained': 0.23, 'Tradition attained': 0, 'Tradition constrained': 0, 'Conformity: rules attained': 0.21, 'Conformity: rules constrained': 0.18, 'Conformity: interpersonal attained': 0, 'Conformity: interpersonal constrained': 0.15, 'Humility attained': 0, 'Humility constrained': 0, 'Benevolence: caring attained': 0.15, 'Benevolence: caring constrained': 0, 'Benevolence: dependability attained': 0, 'Benevolence: dependability constrained': 0, 'Universalism: concern attained': 0.49, 'Universalism: concern constrained': 0.16, 'Universalism: nature attained': 0.44, 'Universalism: nature constrained': 0.12, 'Universalism: tolerance attained': 0, 'Universalism: tolerance constrained': 0.09}
• —Marco-avg-f1-score: 0.12

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
• —trainbatchsize: 8
• —evalbatchsize: 8
• —seed: 42
• —optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• —lrschedulertype: linear
• —num_epochs: 20

Training results

Framework versions

• —Transformers 4.38.2
• —Pytorch 2.2.1+cu121
• —Datasets 2.18.0
• —Tokenizers 0.15.2