TRBLLmaker / README.md
librarian-bot's picture
Librarian Bot: Add base_model information to model
db2862b
|
raw
history blame
2.25 kB
metadata
language:
  - eng
license: afl-3.0
tags:
  - generative-model
datasets:
  - TRBLLmaker
model:
  - gpt2
widget:
  - text: >-
      artist: Ashe.

      title: Moral of the story.

      lyrics: Some mistakes get made That's alright, that's okay You can think
      that you're in love When you're really just in pain.

      meaning:
    example_title: Moral of the story
base_model: gpt2-medium

Link to the arxiv

Datasets

About Us

Created by Mor Ventura and Michael Toker.


tags: - generated_from_trainer metrics: - accuracy model-index: - name: checkpoint_gpt2-medium_lyrics_meaning_2022-03-10-16-16-32 results: []

checkpoint_gpt2-medium_lyrics_meaning_2022-03-10-16-16-32

This model is a fine-tuned version of gpt2-medium on an unknown dataset. It achieves the following results on the evaluation set:

  • Loss: 2.5896
  • Accuracy: 0.4923

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: 2
  • eval_batch_size: 8
  • seed: 42
  • gradient_accumulation_steps: 16
  • total_train_batch_size: 32
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • num_epochs: 4.0
  • mixed_precision_training: Native AMP

Training results

Training Loss Epoch Step Validation Loss Accuracy
2.7781 1.01 128 2.6284 0.4875
2.6217 2.02 256 2.6022 0.4908
2.569 3.02 384 2.5928 0.4917

Framework versions

  • Transformers 4.18.0.dev0
  • Pytorch 1.10.1
  • Datasets 1.18.3
  • Tokenizers 0.11.6