adamjweintraut's picture
Upload tokenizer
c725741 verified
metadata
license: apache-2.0
tags:
  - generated_from_trainer
base_model: facebook/bart-large
model-index:
  - name: bart-finetuned-lyrlen-256-tokens
    results: []

bart-finetuned-lyrlen-256-tokens

This model is a fine-tuned version of facebook/bart-large on an unknown dataset. It achieves the following results on the evaluation set:

  • Loss: 2.6246

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: 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: 4

Training results

Training Loss Epoch Step Validation Loss
2.9847 0.17 500 2.7417
2.7603 0.33 1000 2.6936
2.6777 0.5 1500 2.7032
2.6679 0.67 2000 2.6790
2.631 0.83 2500 2.6584
2.612 1.0 3000 2.6528
2.5602 1.17 3500 2.7752
2.5321 1.33 4000 2.6442
2.523 1.5 4500 2.6168
2.5031 1.67 5000 2.6356
2.4454 1.83 5500 2.6987
2.5141 2.0 6000 2.6277
2.434 2.17 6500 2.6612
2.4381 2.33 7000 2.6339
2.4182 2.5 7500 2.6291
2.3564 2.67 8000 2.6309
2.4002 2.83 8500 2.5957
2.3692 3.0 9000 2.6150
2.4242 3.17 9500 2.5908
2.3854 3.33 10000 2.6053
2.3493 3.5 10500 2.6875
2.3231 3.67 11000 2.6197
2.3609 3.83 11500 2.6016
2.3651 4.0 12000 2.6246

Framework versions

  • Transformers 4.40.0.dev0
  • Pytorch 2.2.1+cu121
  • Datasets 2.18.0
  • Tokenizers 0.15.2