DialoGPT-large-faqs-block-size-128-bs-16-lr-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: 3.3741

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: 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	2.6793
No log	2.0	80	2.3038
No log	3.0	120	2.2566
No log	4.0	160	2.3382
No log	5.0	200	2.5499
No log	6.0	240	2.6927
No log	7.0	280	2.8513
No log	8.0	320	2.9774
No log	9.0	360	3.0255
No log	10.0	400	3.1119
No log	11.0	440	3.1643
No log	12.0	480	3.2005
0.9696	13.0	520	3.2673
0.9696	14.0	560	3.2855
0.9696	15.0	600	3.3351
0.9696	16.0	640	3.3462
0.9696	17.0	680	3.3375
0.9696	18.0	720	3.3614
0.9696	19.0	760	3.3648
0.9696	20.0	800	3.3741

Framework versions

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