2_task_v2_new2labels

This model is a fine-tuned version of ttqdunggg/finetune_cls_vs_content_ronbackbone_100k on the None dataset. It achieves the following results on the evaluation set:

• Loss: nan
• F1 New: 0.7766
• Acc New: 0.7868

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: 32
• seed: 42
• optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: cosine
• lr_scheduler_warmup_steps: 80
• num_epochs: 20
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	F1 New	Acc New
0.6547	1.0	40	nan	0.6830	0.7426
0.5695	2.0	80	nan	0.7176	0.7574
0.5323	3.0	120	nan	0.7236	0.75
0.5104	4.0	160	nan	0.7527	0.7647
0.4958	5.0	200	nan	0.7654	0.7868
0.4795	6.0	240	nan	0.7405	0.75
0.4719	7.0	280	nan	0.7874	0.8015
0.4584	8.0	320	nan	0.7874	0.8015
0.4528	9.0	360	nan	0.7612	0.7721
0.4434	10.0	400	nan	0.7766	0.7868
0.4443	11.0	440	nan	0.7961	0.8088
0.4358	12.0	480	nan	0.7836	0.7941
0.4393	13.0	520	nan	0.7696	0.7794
0.4391	14.0	560	nan	0.7766	0.7868
0.4265	15.0	600	nan	0.7836	0.7941
0.4270	16.0	640	nan	0.7766	0.7868

Framework versions

• Transformers 5.13.1
• Pytorch 2.11.0+cu128
• Datasets 4.0.0
• Tokenizers 0.22.2