DialoGPT-large-faqs-block-size-64-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: 2.7013

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	81	3.6517
No log	2.0	162	2.9681
No log	3.0	243	2.7276
No log	4.0	324	2.6117
No log	5.0	405	2.5552
No log	6.0	486	2.5121
2.8918	7.0	567	2.4886
2.8918	8.0	648	2.4982
2.8918	9.0	729	2.4839
2.8918	10.0	810	2.5141
2.8918	11.0	891	2.5546
2.8918	12.0	972	2.5802
1.4641	13.0	1053	2.6159
1.4641	14.0	1134	2.6419
1.4641	15.0	1215	2.6417
1.4641	16.0	1296	2.6639
1.4641	17.0	1377	2.6681
1.4641	18.0	1458	2.6862
1.0613	19.0	1539	2.7011
1.0613	20.0	1620	2.7013

Framework versions

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