short_first_nodatives_seed-42_1e-3

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

• Loss: 3.2991
• Accuracy: 0.3878

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
6.1422	1.0	1731	4.4278	0.2877
4.1394	2.0	3462	3.9616	0.3216
3.7204	3.0	5193	3.7141	0.3428
3.579	4.0	6924	3.5867	0.3550
3.4012	5.0	8655	3.5119	0.3619
3.3035	6.0	10386	3.4604	0.3676
3.2368	7.0	12117	3.4316	0.3699
3.1867	8.0	13848	3.4069	0.3723
3.148	9.0	15579	3.4014	0.3734
3.1267	10.0	17310	3.3863	0.3748
3.1098	11.0	19041	3.3862	0.3748
3.0629	12.0	20772	3.3839	0.3753
3.059	13.0	22503	3.3726	0.3765
3.0584	14.0	24234	3.3704	0.3776
3.0259	15.0	25965	3.3641	0.3777
3.0138	16.0	27696	3.3705	0.3776
3.0263	17.0	29427	3.3593	0.3780
3.0342	18.0	31158	3.3652	0.3777
3.0019	19.0	32889	3.3312	0.3819
2.8727	20.0	34620	3.2991	0.3878

Framework versions

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