xlm-r-large-phoner

This model is a fine-tuned version of xlm-roberta-large on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.3043
• Micro Precision: 0.9548
• Micro Recall: 0.9610
• Micro F1: 0.9579

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: 16
• eval_batch_size: 16
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 64
• 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: linear
• num_epochs: 30
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Micro Precision	Micro Recall	Micro F1
No log	1.0	79	0.2085	0.8585	0.9376	0.8963
No log	2.0	158	0.1734	0.9036	0.9355	0.9193
No log	3.0	237	0.1471	0.9238	0.9473	0.9354
No log	4.0	316	0.1700	0.9384	0.9492	0.9438
No log	5.0	395	0.1788	0.9213	0.9453	0.9331
No log	6.0	474	0.1763	0.9364	0.9589	0.9475
0.7363	7.0	553	0.1738	0.9410	0.9528	0.9468
0.7363	8.0	632	0.1940	0.9400	0.9589	0.9494
0.7363	9.0	711	0.1892	0.9393	0.9575	0.9483
0.7363	10.0	790	0.2084	0.9423	0.9551	0.9487
0.7363	11.0	869	0.2393	0.9455	0.9536	0.9495
0.7363	12.0	948	0.2102	0.9491	0.9556	0.9524
0.0970	13.0	1027	0.2251	0.9422	0.9543	0.9482
0.0970	14.0	1106	0.2360	0.9443	0.9642	0.9541
0.0970	15.0	1185	0.2485	0.9453	0.9565	0.9509
0.0970	16.0	1264	0.2535	0.9480	0.9561	0.9521
0.0970	17.0	1343	0.2480	0.9464	0.9593	0.9528
0.0970	18.0	1422	0.2899	0.9453	0.9584	0.9518
0.0398	19.0	1501	0.2723	0.9503	0.9596	0.9550
0.0398	20.0	1580	0.2587	0.9482	0.9597	0.9539
0.0398	21.0	1659	0.2986	0.9508	0.9589	0.9549
0.0398	22.0	1738	0.2739	0.9535	0.9615	0.9575
0.0398	23.0	1817	0.2916	0.9524	0.9626	0.9574
0.0398	24.0	1896	0.2959	0.9532	0.9612	0.9572
0.0398	25.0	1975	0.2933	0.9535	0.9596	0.9565
0.0118	26.0	2054	0.2937	0.9541	0.9607	0.9574
0.0118	27.0	2133	0.3011	0.9570	0.9592	0.9581
0.0118	28.0	2212	0.3029	0.9557	0.9601	0.9579
0.0118	29.0	2291	0.3034	0.9543	0.9604	0.9573
0.0118	30.0	2370	0.3043	0.9548	0.9610	0.9579

Framework versions

• Transformers 5.0.0
• Pytorch 2.10.0+cu128
• Datasets 4.8.3
• Tokenizers 0.22.2