xlm-roberta-base-som-finetuned

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

• Loss: 0.3279
• F1: 0.5521
• Roc Auc: 0.7307
• Accuracy: 0.4841

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: 32
• eval_batch_size: 32
• 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_steps: 100
• num_epochs: 20

Training results

Training Loss	Epoch	Step	Validation Loss	F1	Roc Auc	Accuracy
0.3844	1.0	174	0.3804	0.0499	0.5137	0.2954
0.3404	2.0	348	0.3427	0.2255	0.5749	0.3379
0.3261	3.0	522	0.3101	0.3609	0.6337	0.4215
0.2787	4.0	696	0.3072	0.3492	0.6277	0.4143
0.2652	5.0	870	0.3143	0.4482	0.6769	0.4467
0.2333	6.0	1044	0.2970	0.4876	0.6852	0.4654
0.2169	7.0	1218	0.3066	0.4911	0.6937	0.4633
0.1875	8.0	1392	0.3103	0.5438	0.7236	0.4813
0.1747	9.0	1566	0.3341	0.5215	0.7157	0.4820
0.1611	10.0	1740	0.3281	0.5332	0.7232	0.4741
0.1371	11.0	1914	0.3279	0.5521	0.7307	0.4841
0.1245	12.0	2088	0.3340	0.5415	0.7246	0.4777
0.1197	13.0	2262	0.3462	0.5388	0.7238	0.4755
0.0978	14.0	2436	0.3530	0.5408	0.7268	0.4733
0.0977	15.0	2610	0.3543	0.5377	0.7250	0.4705

Framework versions

• Transformers 4.47.1
• Pytorch 2.5.1+cu121
• Datasets 3.2.0
• Tokenizers 0.21.0