roberta-base_lora_lr5e-05_bs4_epoch20_wd0.01

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

• Loss: 0.3786

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: 4
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 500
• num_epochs: 20

Training results

Training Loss	Epoch	Step	Validation Loss
19.0917	1.0	157	12.3781
5.0512	2.0	314	4.0762
3.8203	3.0	471	2.6873
2.586	4.0	628	2.0339
2.2522	5.0	785	1.6467
1.8436	6.0	942	1.2936
1.7033	7.0	1099	1.0691
1.4103	8.0	1256	0.8708
1.2267	9.0	1413	0.7359
1.1845	10.0	1570	0.6333
1.0321	11.0	1727	0.5510
0.9554	12.0	1884	0.5251
0.897	13.0	2041	0.4977
0.8424	14.0	2198	0.4571
0.8221	15.0	2355	0.4298
0.7638	16.0	2512	0.4095
0.7494	17.0	2669	0.4057
0.751	18.0	2826	0.3920
0.7365	19.0	2983	0.3818
0.7312	20.0	3140	0.3786

Framework versions

• PEFT 0.7.1
• Transformers 4.36.2
• Pytorch 2.0.1
• Datasets 2.16.1
• Tokenizers 0.15.0