candor_o_67

This model is a fine-tuned version of on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 3.4942

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: 0.0001
• train_batch_size: 256
• eval_batch_size: 256
• seed: 67
• 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: 500
• num_epochs: 20
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss
5.2748	1.0	422	4.0635
3.7636	2.0	844	3.7294
3.5487	3.0	1266	3.6128
3.4377	4.0	1688	3.5462
3.3589	5.0	2110	3.5060
3.2954	6.0	2532	3.4735
3.2399	7.0	2954	3.4535
3.1898	8.0	3376	3.4374
3.1427	9.0	3798	3.4312
3.0978	10.0	4220	3.4254
3.0538	11.0	4642	3.4280
3.0119	12.0	5064	3.4306
2.9709	13.0	5486	3.4365
2.932	14.0	5908	3.4458
2.894	15.0	6330	3.4547
2.8597	16.0	6752	3.4645
2.8297	17.0	7174	3.4755
2.8037	18.0	7596	3.4844
2.7826	19.0	8018	3.4903
2.7657	20.0	8440	3.4942

Framework versions

• Transformers 4.56.1
• Pytorch 2.8.0+cu128
• Datasets 4.0.0
• Tokenizers 0.22.0