DrishtiSharma's picture
End of training
1ceca7c
metadata
license: mit
base_model: microsoft/DialoGPT-large
tags:
  - generated_from_trainer
model-index:
  - name: DialoGPT-large-faqs-block-size-128-bs-16-lr-1e-6
    results: []

DialoGPT-large-faqs-block-size-128-bs-16-lr-1e-6

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: 3.9454

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-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 5.9267
No log 2.0 80 5.4947
No log 3.0 120 5.2047
No log 4.0 160 4.9717
No log 5.0 200 4.7953
No log 6.0 240 4.6423
No log 7.0 280 4.5234
No log 8.0 320 4.4199
No log 9.0 360 4.3313
No log 10.0 400 4.2584
No log 11.0 440 4.1938
No log 12.0 480 4.1346
5.1633 13.0 520 4.0863
5.1633 14.0 560 4.0490
5.1633 15.0 600 4.0146
5.1633 16.0 640 3.9882
5.1633 17.0 680 3.9686
5.1633 18.0 720 3.9556
5.1633 19.0 760 3.9480
5.1633 20.0 800 3.9454

Framework versions

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