metadata
license: apache-2.0
base_model: google/flan-t5-small
tags:
- generated_from_trainer
datasets:
- race
model-index:
- name: flan-small-mc-question-options-generation
results: []
inference:
parameters:
max_length: 512
num_beams: 4
length_penalty: 1.5
no_repeat_ngram_size: 3
early_stopping: true
flan-small-mc-question-options-generation
This model is a fine-tuned version of google/flan-t5-small on the race dataset. It achieves the following results on the evaluation set:
- Loss: 1.5747
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: 4
- eval_batch_size: 4
- seed: 42
- gradient_accumulation_steps: 16
- total_train_batch_size: 64
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 3
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 4.3811 | 0.07 | 100 | 2.4007 |
| 2.5622 | 0.15 | 200 | 2.0431 |
| 2.2498 | 0.22 | 300 | 1.9139 |
| 2.1136 | 0.29 | 400 | 1.8218 |
| 2.0392 | 0.36 | 500 | 1.7739 |
| 1.9984 | 0.44 | 600 | 1.7237 |
| 1.9519 | 0.51 | 700 | 1.7058 |
| 1.928 | 0.58 | 800 | 1.6900 |
| 1.9075 | 0.66 | 900 | 1.6687 |
| 1.8968 | 0.73 | 1000 | 1.6631 |
| 1.8758 | 0.8 | 1100 | 1.6519 |
| 1.8863 | 0.87 | 1200 | 1.6431 |
| 1.869 | 0.95 | 1300 | 1.6409 |
| 1.8483 | 1.02 | 1400 | 1.6313 |
| 1.8344 | 1.09 | 1500 | 1.6279 |
| 1.8398 | 1.17 | 1600 | 1.6183 |
| 1.8247 | 1.24 | 1700 | 1.6223 |
| 1.8131 | 1.31 | 1800 | 1.6072 |
| 1.8024 | 1.38 | 1900 | 1.6096 |
| 1.8038 | 1.46 | 2000 | 1.6056 |
| 1.8051 | 1.53 | 2100 | 1.6030 |
| 1.7875 | 1.6 | 2200 | 1.6008 |
| 1.7983 | 1.68 | 2300 | 1.5923 |
| 1.7922 | 1.75 | 2400 | 1.5917 |
| 1.7892 | 1.82 | 2500 | 1.5903 |
| 1.784 | 1.89 | 2600 | 1.5891 |
| 1.7844 | 1.97 | 2700 | 1.5867 |
| 1.7678 | 2.04 | 2800 | 1.5837 |
| 1.7558 | 2.11 | 2900 | 1.5826 |
| 1.7702 | 2.19 | 3000 | 1.5820 |
| 1.7669 | 2.26 | 3100 | 1.5802 |
| 1.7715 | 2.33 | 3200 | 1.5763 |
| 1.7724 | 2.4 | 3300 | 1.5799 |
| 1.757 | 2.48 | 3400 | 1.5783 |
| 1.7648 | 2.55 | 3500 | 1.5763 |
| 1.7691 | 2.62 | 3600 | 1.5772 |
| 1.7607 | 2.69 | 3700 | 1.5758 |
| 1.7574 | 2.77 | 3800 | 1.5745 |
| 1.7586 | 2.84 | 3900 | 1.5748 |
| 1.7583 | 2.91 | 4000 | 1.5745 |
| 1.7611 | 2.99 | 4100 | 1.5747 |
Framework versions
- Transformers 4.31.0
- Pytorch 2.0.0
- Datasets 2.1.0
- Tokenizers 0.13.3