DialoGPT-large-faqs-block-size128-bs-16

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: 3.5086

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: 7e-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	2.4979
No log	2.0	80	2.2314
No log	3.0	120	2.2409
No log	4.0	160	2.4555
No log	5.0	200	2.7390
No log	6.0	240	2.9258
No log	7.0	280	3.0355
No log	8.0	320	3.1368
No log	9.0	360	3.2088
No log	10.0	400	3.2541
No log	11.0	440	3.3225
No log	12.0	480	3.3775
0.7809	13.0	520	3.4102
0.7809	14.0	560	3.4456
0.7809	15.0	600	3.4707
0.7809	16.0	640	3.4786
0.7809	17.0	680	3.4868
0.7809	18.0	720	3.5035
0.7809	19.0	760	3.5015
0.7809	20.0	800	3.5086

Framework versions

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