DialoGPT-large-faqs-block-size-128-bs-16-lr-0.5e-5

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

• Loss: 2.5447

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-06
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 20

Training results

Training Loss	Epoch	Step	Validation Loss
No log	1.0	40	4.7556
No log	2.0	80	4.0984
No log	3.0	120	3.6525
No log	4.0	160	3.3247
No log	5.0	200	3.1137
No log	6.0	240	2.9706
No log	7.0	280	2.8696
No log	8.0	320	2.7942
No log	9.0	360	2.7382
No log	10.0	400	2.6928
No log	11.0	440	2.6547
No log	12.0	480	2.6237
3.3313	13.0	520	2.6033
3.3313	14.0	560	2.5852
3.3313	15.0	600	2.5690
3.3313	16.0	640	2.5614
3.3313	17.0	680	2.5532
3.3313	18.0	720	2.5485
3.3313	19.0	760	2.5458
3.3313	20.0	800	2.5447

Framework versions

• Transformers 4.33.0.dev0
• Pytorch 2.0.1+cu118
• Datasets 2.14.4.dev0
• Tokenizers 0.13.3