roberta-base_lora_lr0.0005_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: 0.0056

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: 0.0005
• 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
12.0309	1.0	157	2.8075
2.0953	2.0	314	1.1630
1.3141	3.0	471	0.3769
0.5659	4.0	628	0.1709
0.3778	5.0	785	0.1162
0.2225	6.0	942	0.0794
0.1871	7.0	1099	0.0458
0.1331	8.0	1256	0.0276
0.0819	9.0	1413	0.0264
0.0803	10.0	1570	0.0166
0.057	11.0	1727	0.0161
0.0525	12.0	1884	0.0178
0.0378	13.0	2041	0.0099
0.0357	14.0	2198	0.0051
0.0248	15.0	2355	0.0077
0.0292	16.0	2512	0.0047
0.0229	17.0	2669	0.0052
0.0227	18.0	2826	0.0062
0.0201	19.0	2983	0.0056
0.0198	20.0	3140	0.0056

Framework versions

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