aha_sentence_classification

This model is a fine-tuned version of skt/A.X-Encoder-base on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.8454
• Accuracy: 0.6900
• F1 Micro: 0.6900
• F1 Macro: 0.6503
• Precision Macro: 0.6078
• Recall Macro: 0.7221

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: 64
• eval_batch_size: 64
• seed: 42
• optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: cosine
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 25

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1 Micro	F1 Macro	Precision Macro	Recall Macro
0.9702	0.5949	1000	1.1520	0.5590	0.5590	0.5444	0.5142	0.6791
0.7293	1.1898	2000	1.0469	0.5992	0.5992	0.5966	0.5599	0.7238
0.7779	1.7847	3000	0.9977	0.6278	0.6278	0.5964	0.5646	0.7274
0.5545	2.3795	4000	0.9847	0.6290	0.6290	0.6208	0.5849	0.7236
0.5692	2.9744	5000	0.8454	0.6900	0.6900	0.6503	0.6078	0.7221
0.3962	3.5693	6000	1.0074	0.6488	0.6488	0.6316	0.6093	0.7081
0.1624	4.1642	7000	1.1059	0.6732	0.6732	0.6533	0.6322	0.6930
0.1816	4.7591	8000	1.1277	0.6872	0.6872	0.6513	0.6429	0.6690
0.0934	5.3540	9000	1.4084	0.6882	0.6882	0.6468	0.6380	0.6649
0.0882	5.9488	10000	1.4941	0.6918	0.6918	0.6450	0.6428	0.6606

Framework versions

• Transformers 4.56.1
• Pytorch 2.7.0+cu126
• Tokenizers 0.22.0