datives_removed_seed-42_1e-3

This model was trained from scratch on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 3.1613
• Accuracy: 0.4025

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: 0.001
• train_batch_size: 32
• eval_batch_size: 64
• seed: 42
• gradient_accumulation_steps: 8
• total_train_batch_size: 256
• 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
• lr_scheduler_warmup_steps: 32000
• num_epochs: 20.0
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
6.0296	1.0	1523	4.3935	0.2939
3.9478	2.0	3046	3.8752	0.3349
3.6967	3.0	4569	3.5985	0.3586
3.4052	4.0	6092	3.4406	0.3733
3.2986	5.0	7615	3.3454	0.3821
3.1794	6.0	9138	3.2891	0.3877
3.1191	7.0	10661	3.2542	0.3908
3.0619	8.0	12184	3.2253	0.3940
3.0169	9.0	13707	3.2092	0.3959
2.9884	10.0	15230	3.1991	0.3974
2.9541	11.0	16753	3.1869	0.3983
2.9379	12.0	18276	3.1832	0.3991
2.9128	13.0	19799	3.1771	0.4000
2.9004	14.0	21322	3.1749	0.4006
2.8831	15.0	22845	3.1711	0.4007
2.8732	16.0	24368	3.1699	0.4012
2.8658	17.0	25891	3.1691	0.4016
2.855	18.0	27414	3.1670	0.4016
2.8498	19.0	28937	3.1642	0.4018
2.8398	20.0	30460	3.1613	0.4025

Framework versions

• Transformers 4.46.2
• Pytorch 2.5.1+cu124
• Datasets 3.2.0
• Tokenizers 0.20.0