train_4

This model was trained from scratch on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.4761

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: 5e-05
• train_batch_size: 128
• eval_batch_size: 128
• seed: 42
• optimizer: Use 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: 5000
• num_epochs: 200

Training results

Training Loss	Epoch	Step	Validation Loss
0.0312	1.0	8505	0.3723
0.0305	2.0	17010	0.3694
0.0304	3.0	25515	0.3974
0.03	4.0	34020	0.3869
0.0297	5.0	42525	0.4023
0.029	6.0	51030	0.3562
0.0289	7.0	59535	0.3798
0.0285	8.0	68040	0.3982
0.028	9.0	76545	0.3601
0.028	10.0	85050	0.3971
0.0276	11.0	93555	0.3640
0.0271	12.0	102060	0.3709
0.027	13.0	110565	0.5397
0.0265	14.0	119070	0.3994
0.0264	15.0	127575	0.4113
0.0263	16.0	136080	0.3864
0.0258	17.0	144585	0.5371
0.0257	18.0	153090	0.4825
0.0254	19.0	161595	0.3586
0.025	20.0	170100	0.4152
0.025	21.0	178605	0.4761

Framework versions

• Transformers 4.51.3
• Pytorch 2.8.0.dev20250429
• Datasets 2.14.0
• Tokenizers 0.21.1