modernbert_seeker1

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

• Loss: 3.6513
• F1: 0.6676

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: 4
• eval_batch_size: 4
• 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
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 25

Training results

Training Loss	Epoch	Step	Validation Loss	F1
No log	1.0	92	1.1453	0.2556
1.2432	2.0	184	1.1348	0.2269
1.1447	3.0	276	1.1944	0.2624
1.0924	4.0	368	0.9543	0.4673
0.9918	5.0	460	1.0970	0.4892
0.8981	6.0	552	1.2106	0.5669
0.9261	7.0	644	0.8952	0.6269
0.8208	8.0	736	1.1409	0.5480
0.645	9.0	828	1.7286	0.5748
0.6745	10.0	920	1.4750	0.6229
0.5947	11.0	1012	1.7091	0.6628
0.517	12.0	1104	3.0948	0.6180
0.517	13.0	1196	2.5940	0.6671
0.4901	14.0	1288	2.9827	0.6348
0.2048	15.0	1380	3.1023	0.6483
0.1188	16.0	1472	3.7382	0.6332
0.1236	17.0	1564	3.5499	0.6535
0.0828	18.0	1656	3.6281	0.6533
0.0479	19.0	1748	3.2783	0.6820
0.0081	20.0	1840	3.6055	0.6676
0.031	21.0	1932	3.6255	0.6778
0.0008	22.0	2024	3.6279	0.6850
0.0	23.0	2116	3.6376	0.6676
0.0001	24.0	2208	3.6440	0.6676
0.0	25.0	2300	3.6513	0.6676

Framework versions

• Transformers 4.48.0
• Pytorch 2.5.1+cu121
• Datasets 3.2.0
• Tokenizers 0.21.0