mrm8488's picture
Update README.md
0542559
metadata
license: bigscience-bloom-rail-1.0
langs:
  - es
tags:
  - generated_from_trainer
  - summarization
  - Spanish summarizing
model-index:
  - name: bloom-560m-finetuned-wikilingua-spanish-summarization
    results: []

bloom-560m-finetuned-wikilingua-spanish-summarization

This model is a fine-tuned version of bigscience/bloom-560m on the None dataset. It achieves the following results on the evaluation set:

  • Loss: 2.1223

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: 1
  • eval_batch_size: 2
  • seed: 42
  • gradient_accumulation_steps: 4
  • total_train_batch_size: 4
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • num_epochs: 2
  • mixed_precision_training: Native AMP

Training results

Training Loss Epoch Step Validation Loss
2.4768 0.07 500 2.4828
2.428 0.14 1000 2.4125
2.4 0.2 1500 2.3927
2.3685 0.27 2000 2.3506
2.3287 0.34 2500 2.3340
2.3196 0.41 3000 2.3284
2.2885 0.48 3500 2.3005
2.2646 0.55 4000 2.2944
2.2676 0.68 5000 2.2575
2.2267 0.82 6000 2.2281
2.1971 0.95 7000 2.2018
2.009 1.09 8000 2.1925
1.9989 1.23 9000 2.1765
2.0131 1.36 10000 2.1666
1.9765 1.5 11000 2.1514
1.9449 1.64 12000 2.1404
1.9399 1.77 13000 2.1297
1.957 1.91 14000 2.1223

Framework versions

  • Transformers 4.23.0.dev0
  • Pytorch 1.12.1+cu113
  • Datasets 2.5.1
  • Tokenizers 0.13.0