metadata
license: apache-2.0
base_model: google/flan-t5-base
tags:
- generated_from_trainer
metrics:
- rouge
model-index:
- name: GK_10000
results: []
datasets:
- MuskumPillerum/General-Knowledge
GK_10000
This model is a fine-tuned version of google/flan-t5-base on the MuskumPillerum/General-Knowledge dataset. It achieves the following results on the evaluation set:
- Loss: 1.6340
- Rouge1: 0.4195
- Rouge2: 0.3016
- Rougel: 0.3924
- Rougelsum: 0.3995
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: 0.0003
- train_batch_size: 8
- eval_batch_size: 4
- seed: 42
- 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 | Rouge1 | Rouge2 | Rougel | Rougelsum |
|---|---|---|---|---|---|---|---|
| 2.1243 | 1.0 | 875 | 1.7023 | 0.4106 | 0.2880 | 0.3827 | 0.3899 |
| 1.6975 | 2.0 | 1750 | 1.6422 | 0.4231 | 0.3026 | 0.3957 | 0.4025 |
| 1.495 | 3.0 | 2625 | 1.6340 | 0.4195 | 0.3016 | 0.3924 | 0.3995 |
Framework versions
- Transformers 4.42.4
- Pytorch 2.3.1+cu121
- Datasets 2.20.0
- Tokenizers 0.19.1