gs-aristoBERTo

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

• Loss: 1.4418
• Top1: 23.4940
• Top5: 44.7791
• Top10: 54.0161
• Top20: 56.8273
• Bertscore F1 Top1: 82.2464
• Bertscore F1 Top1 Mean: 82.2464
• Bertscore F1 Top5: 88.2069
• Bertscore F1 Top5 Mean: 79.8852
• Bertscore F1 Top10: 90.2564
• Bertscore F1 Top10 Mean: 78.7899
• Bertscore F1 Top20: 91.1153
• Bertscore F1 Top20 Mean: 77.5504
• Cos Sim Top1 Max: 71.0649
• Cos Sim Top1 Mean: 71.0649
• Cos Sim Top5 Max: 82.3444
• Cos Sim Top5 Mean: 67.2349
• Cos Sim Top10 Max: 86.3235
• Cos Sim Top10 Mean: 65.5829
• Cos Sim Top20 Max: 88.1500
• Cos Sim Top20 Mean: 64.2762
• Cluster Inclusion Rate: 64.2570
• Mean Inclusion Margin: 0.0577
• Mean Gold Centroid Cosine Sim: 76.7035

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: 4.192159350410978e-06
• train_batch_size: 64
• eval_batch_size: 64
• seed: 42
• optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED 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: 0.1
• num_epochs: 3

Training results

Training Loss	Epoch	Step	Validation Loss	Top1	Top5	Top10	Top20	Bertscore F1 Top1	Bertscore F1 Top1 Mean	Bertscore F1 Top5	Bertscore F1 Top5 Mean	Bertscore F1 Top10	Bertscore F1 Top10 Mean	Bertscore F1 Top20	Bertscore F1 Top20 Mean	Cos Sim Top1 Max	Cos Sim Top1 Mean	Cos Sim Top5 Max	Cos Sim Top5 Mean	Cos Sim Top10 Max	Cos Sim Top10 Mean	Cos Sim Top20 Max	Cos Sim Top20 Mean	Cluster Inclusion Rate	Mean Inclusion Margin	Mean Gold Centroid Cosine Sim
1.5818	1.0	385	1.4636	23.6948	44.7791	54.6185	57.4297	82.2605	82.2605	88.1323	79.8213	90.2998	78.7970	91.2274	77.6046	71.0099	71.0099	82.3104	67.3073	86.4382	65.8246	88.2805	64.5302	63.8554	0.0587	76.8632
1.5725	2.0	770	1.4538	23.4940	44.7791	54.0161	56.8273	82.2464	82.2464	88.2069	79.8852	90.2564	78.7899	91.1153	77.5504	71.0649	71.0649	82.3444	67.2349	86.3235	65.5829	88.1500	64.2762	64.2570	0.0577	76.7035
1.5541	3.0	1155	1.4543	23.2932	45.3815	54.4177	57.0281	82.3101	82.3101	88.3667	79.9315	90.3639	78.7954	91.2038	77.5678	71.0135	71.0135	82.4616	67.3003	86.4104	65.6693	88.1290	64.3142	64.2570	0.0588	76.7485

Framework versions

• Transformers 5.17.0
• Pytorch 2.14.0+cu126
• Datasets 3.6.0
• Tokenizers 0.23.2