loose_balanced_seed-42_1e-3

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

• Loss: 3.0214
• Accuracy: 0.4202

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
6.1274	0.9998	1776	4.2576	0.3052
4.0402	1.9996	3552	3.7328	0.3476
3.5742	2.9995	5328	3.4742	0.3713
3.3401	3.9999	7105	3.3362	0.3841
3.2496	4.9998	8881	3.2569	0.3914
3.1455	5.9996	10657	3.2070	0.3963
3.0827	6.9995	12433	3.1771	0.3992
3.0395	7.9999	14210	3.1554	0.4018
2.9985	8.9998	15986	3.1429	0.4034
2.9602	9.9996	17762	3.1328	0.4041
2.9492	10.9995	19538	3.1212	0.4055
2.9388	11.9999	21315	3.1175	0.4062
2.9335	12.9998	23091	3.1109	0.4074
2.8912	13.9996	24867	3.1067	0.4077
2.89	14.9995	26643	3.1065	0.4077
2.894	15.9999	28420	3.1024	0.4082
2.8975	16.9998	30196	3.0992	0.4086
2.8675	17.9996	31972	3.0961	0.4088
2.8388	18.9995	33748	3.0460	0.4155
2.6928	19.9971	35520	3.0214	0.4202

Framework versions

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