long_first_headfinal_seed-63_1e-3

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

• Loss: 5.1385
• Accuracy: 0.1992

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: 63
• 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
6.136	0.9994	1470	5.5155	0.1759
4.5192	1.9992	2940	5.3994	0.1849
3.8912	2.9991	4410	5.2905	0.1867
3.7139	3.9996	5881	5.2149	0.1951
3.4884	4.9994	7351	5.1594	0.1973
3.4101	5.9992	8821	5.1296	0.2025
3.3062	6.9991	10291	5.1459	0.1937
3.2596	7.9996	11762	5.1509	0.1967
3.2055	8.9994	13232	5.0996	0.2014
3.1686	9.9992	14702	5.1043	0.2000
3.1453	10.9991	16172	5.0935	0.2006
3.1092	11.9996	17643	5.0770	0.2010
3.1047	12.9994	19113	5.0922	0.2011
3.0688	13.9992	20583	5.0953	0.2020
3.0784	14.9991	22053	5.1038	0.2010
3.0404	15.9996	23524	5.0954	0.2020
3.0417	16.9994	24994	5.1111	0.1999
3.0187	17.9992	26464	5.1096	0.2008
3.0244	18.9991	27934	5.0990	0.2001
3.0117	19.9962	29400	5.1385	0.1992

Framework versions

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