DialoGPT-large-faqs-block-size-400-bs-16-lr-1e-05

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: 2.6788

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: 1e-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	13	5.8404
No log	2.0	26	5.0890
No log	3.0	39	4.6086
No log	4.0	52	4.1995
No log	5.0	65	3.8393
No log	6.0	78	3.5620
No log	7.0	91	3.3420
No log	8.0	104	3.1838
No log	9.0	117	3.0612
No log	10.0	130	2.9650
No log	11.0	143	2.8926
No log	12.0	156	2.8375
No log	13.0	169	2.8019
No log	14.0	182	2.7654
No log	15.0	195	2.7348
No log	16.0	208	2.7130
No log	17.0	221	2.6977
No log	18.0	234	2.6874
No log	19.0	247	2.6813
No log	20.0	260	2.6788

Framework versions

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