distilroberta-pr200k-ep20

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

• Loss: 1.0749

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: 7.3e-05
• train_batch_size: 64
• eval_batch_size: 64
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.19
• num_epochs: 20

Training results

Training Loss	Epoch	Step	Validation Loss
1.8844	1.0	5480	1.6711
1.6262	2.0	10960	1.4794
1.5005	3.0	16440	1.3796
1.4432	4.0	21920	1.3313
1.3904	5.0	27400	1.2843
1.3378	6.0	32880	1.2441
1.3057	7.0	38360	1.2149
1.2781	8.0	43840	1.1927
1.2497	9.0	49320	1.1780
1.2219	10.0	54800	1.1626
1.2068	11.0	60280	1.1500
1.1832	12.0	65760	1.1368
1.1789	13.0	71240	1.1230
1.1571	14.0	76720	1.1135
1.1439	15.0	82200	1.1020
1.1419	16.0	87680	1.0954
1.1247	17.0	93160	1.0893
1.1159	18.0	98640	1.0806
1.1119	19.0	104120	1.0776
1.0978	20.0	109600	1.0740

Framework versions

• Transformers 4.33.2
• Pytorch 2.0.1+cu118
• Datasets 2.14.5
• Tokenizers 0.13.3