modernbert_seeker

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: 1.8876
• F1: 0.7635

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: 5e-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
• num_epochs: 25

Training results

Training Loss	Epoch	Step	Validation Loss	F1
No log	1.0	92	0.6179	0.5993
0.7932	2.0	184	0.6113	0.5993
0.6589	3.0	276	0.6632	0.6305
0.6418	4.0	368	0.6268	0.5993
0.6315	5.0	460	0.9375	0.2508
0.6522	6.0	552	0.5820	0.5886
0.6183	7.0	644	0.5538	0.6857
0.6136	8.0	736	0.5223	0.6736
0.496	9.0	828	0.7308	0.7777
0.4858	10.0	920	0.7452	0.7832
0.4181	11.0	1012	0.7523	0.7799
0.3395	12.0	1104	1.1841	0.7417
0.3395	13.0	1196	0.7770	0.8181
0.2375	14.0	1288	1.1497	0.7773
0.1728	15.0	1380	1.5189	0.7635
0.0324	16.0	1472	1.6500	0.7420
0.0764	17.0	1564	1.5011	0.7635
0.0244	18.0	1656	1.5790	0.7544
0.0002	19.0	1748	1.9299	0.7539
0.0004	20.0	1840	1.7885	0.7635
0.0001	21.0	1932	1.8230	0.7635
0.0001	22.0	2024	1.8578	0.7635
0.0	23.0	2116	1.8765	0.7635
0.0	24.0	2208	1.8845	0.7635
0.0	25.0	2300	1.8876	0.7635

Framework versions

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