5e-6_xlm-R-xl_Conspiracy_training_with_callbacks

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

• Loss: 0.0624
• Macro F1: 0.9918
• Micro F1: 0.9919
• Accuracy: 0.9919

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-06
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 20

Training results

Training Loss	Epoch	Step	Validation Loss	Macro F1	Micro F1	Accuracy
0.6122	1.0	502	0.2712	0.9549	0.9559	0.9559
0.1503	2.0	1004	0.1047	0.9741	0.9744	0.9744
0.0508	3.0	1506	0.0734	0.9835	0.9837	0.9837
0.0226	4.0	2008	0.0837	0.9811	0.9814	0.9814
0.014	5.0	2510	0.0562	0.9882	0.9884	0.9884
0.0014	6.0	3012	0.0514	0.9894	0.9895	0.9895
0.0016	7.0	3514	0.0501	0.9918	0.9919	0.9919
0.0002	8.0	4016	0.0554	0.9918	0.9919	0.9919
0.0001	9.0	4518	0.0607	0.9906	0.9907	0.9907
0.0001	10.0	5020	0.0856	0.9859	0.9861	0.9861
0.0143	11.0	5522	0.0377	0.9929	0.9930	0.9930
0.0001	12.0	6024	0.0538	0.9918	0.9919	0.9919
0.0001	13.0	6526	0.0568	0.9918	0.9919	0.9919
0.0012	14.0	7028	0.0582	0.9918	0.9919	0.9919
0.0	15.0	7530	0.0500	0.9929	0.9930	0.9930
0.0001	16.0	8032	0.0649	0.9918	0.9919	0.9919
0.0	17.0	8534	0.0649	0.9918	0.9919	0.9919
0.0	18.0	9036	0.0648	0.9918	0.9919	0.9919
0.0009	19.0	9538	0.0621	0.9918	0.9919	0.9919
0.0	20.0	10040	0.0624	0.9918	0.9919	0.9919

Framework versions

• Transformers 4.40.2
• Pytorch 2.2.2+cu121
• Datasets 2.18.0
• Tokenizers 0.19.1