README.md

metadata

library_name: transformers
license: apache-2.0
base_model: answerdotai/ModernBERT-base
tags:
  - generated_from_trainer
metrics:
  - f1
  - accuracy
model-index:
  - name: modernbert-tr-classifier
    results: []

modernbert-tr-classifier

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: 0.8071
• F1: 0.7980
• Accuracy: 0.7959

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: 8e-05
• train_batch_size: 128
• eval_batch_size: 128
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 256
• optimizer: Use adamw_torch with betas=(0.9,0.98) and epsilon=1e-06 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: cosine
• lr_scheduler_warmup_steps: 500
• num_epochs: 20

Training results

Training Loss	Epoch	Step	Validation Loss	F1	Accuracy
3.981	1.0	19	1.9665	0.0904	0.1633
3.6781	2.0	38	1.8383	0.1882	0.2449
3.1283	3.0	57	1.4232	0.4896	0.5061
2.4044	4.0	76	1.2398	0.5488	0.5673
1.9166	5.0	95	1.1468	0.5726	0.6041
1.7122	6.0	114	1.0013	0.6649	0.6653
1.4626	7.0	133	0.8954	0.7198	0.7224
1.2173	8.0	152	0.7306	0.7611	0.7592
1.0648	9.0	171	0.7449	0.7412	0.7388
0.9008	10.0	190	0.6874	0.7754	0.7714
0.856	11.0	209	0.6584	0.8071	0.8082
0.7557	12.0	228	0.6046	0.7854	0.7837
0.472	13.0	247	0.8246	0.7428	0.7429
0.4386	14.0	266	0.7892	0.8042	0.8082
0.3418	15.0	285	0.6727	0.8248	0.8286
0.2662	16.0	304	0.8244	0.8144	0.8163
0.1774	17.0	323	0.7832	0.8083	0.8041
0.1246	18.0	342	0.5501	0.8703	0.8694
0.121	18.9730	360	0.8071	0.7980	0.7959

Framework versions

• Transformers 4.48.0
• Pytorch 2.5.1+cu118
• Datasets 3.1.0
• Tokenizers 0.21.0