jackmedda/answerdotai-ModernBERT-base_finetuned_augmented_augmented_chatgpt4

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: 0.6009
• Accuracy: 0.8529
• F1: 0.9123
• Precision: 0.8387
• Recall: 1.0

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
• num_epochs: 50

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1	Precision	Recall
0.7541	1.0	46	1.0379	0.7	0.8235	0.7	1.0
0.6904	2.0	92	0.6320	0.7	0.8235	0.7	1.0
0.5034	3.0	138	0.5816	0.7	0.8	0.75	0.8571
0.4788	4.0	184	0.6822	0.7	0.8235	0.7	1.0
0.4101	5.0	230	1.0326	0.7	0.8235	0.7	1.0
0.3942	6.0	276	0.9368	0.7	0.8235	0.7	1.0
0.3482	7.0	322	0.7998	0.7	0.8235	0.7	1.0
0.5565	8.0	368	1.0760	0.7	0.8235	0.7	1.0
0.2264	9.0	414	1.0318	0.8	0.875	0.7778	1.0
0.3199	10.0	460	0.6990	0.7	0.8	0.75	0.8571
0.6271	11.0	506	1.3217	0.7	0.8235	0.7	1.0
0.5053	12.0	552	2.2078	0.7	0.8235	0.7	1.0
0.4181	13.0	598	1.9610	0.8	0.875	0.7778	1.0
0.0041	14.0	644	2.4483	0.8	0.875	0.7778	1.0
0.1389	15.0	690	1.2595	0.8	0.875	0.7778	1.0
0.1238	16.0	736	1.8522	0.8	0.875	0.7778	1.0
0.003	17.0	782	1.9726	0.8	0.875	0.7778	1.0

Framework versions

• Transformers 4.48.3
• Pytorch 2.3.0+cu121
• Datasets 3.2.0
• Tokenizers 0.21.0