slac-single-head

This model is a fine-tuned version of on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.5482
• F1 Macro: 0.8380
• Precision Macro: 0.8096
• Recall Macro: 0.8688
• F1 Micro: 0.8552
• Precision Micro: 0.8252
• Recall Micro: 0.8874

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 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_steps: 212
• num_epochs: 15

Training results

Training Loss	Epoch	Step	Validation Loss	F1 Macro	Precision Macro	Recall Macro	F1 Micro	Precision Micro	Recall Micro
0.6444	1.0	213	0.5045	0.6683	0.5514	0.9416	0.6611	0.5087	0.9437
0.3875	2.0	426	0.3121	0.8016	0.7045	0.9342	0.8214	0.7272	0.9437
0.292	3.0	639	0.3003	0.8095	0.7256	0.9294	0.8265	0.7398	0.9361
0.2172	4.0	852	0.3231	0.8340	0.7807	0.8982	0.8509	0.7973	0.9122
0.1935	5.0	1065	0.3262	0.8262	0.7628	0.9073	0.8445	0.7826	0.9170
0.154	6.0	1278	0.3807	0.8351	0.7975	0.8794	0.8506	0.8183	0.8855
0.1007	7.0	1491	0.4045	0.8297	0.7774	0.8922	0.8456	0.7902	0.9094
0.0866	8.0	1704	0.4100	0.8289	0.7706	0.9010	0.8434	0.7863	0.9094
0.0671	9.0	1917	0.4667	0.8335	0.7981	0.8726	0.8497	0.8127	0.8903
0.0544	10.0	2130	0.5062	0.8412	0.8139	0.8707	0.8557	0.8254	0.8884
0.0482	11.0	2343	0.5141	0.8335	0.8076	0.8616	0.8521	0.8287	0.8769
0.0377	12.0	2556	0.5217	0.8346	0.8022	0.8699	0.8520	0.8194	0.8874
0.0304	13.0	2769	0.5419	0.8370	0.8104	0.8658	0.8537	0.8266	0.8826
0.0307	14.0	2982	0.5397	0.8367	0.8043	0.8721	0.8533	0.8210	0.8884
0.0238	15.0	3195	0.5482	0.8380	0.8096	0.8688	0.8552	0.8252	0.8874

Framework versions

• Transformers 4.47.0
• Pytorch 2.5.1+cu121
• Datasets 3.3.1
• Tokenizers 0.21.0