metadata
license: apache-2.0
base_model: studio-ousia/mluke-base
tags:
- generated_from_trainer
model-index:
- name: exp_number1
results: []
exp_number1
This model is a fine-tuned version of studio-ousia/mluke-base on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.1505
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: 8
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 16
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 10
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 0.4113 | 1.0 | 221 | 0.4158 |
| 0.3323 | 2.0 | 443 | 0.2946 |
| 0.2299 | 3.0 | 664 | 0.2595 |
| 0.2158 | 4.0 | 886 | 0.2567 |
| 0.1404 | 5.0 | 1107 | 0.2323 |
| 0.1266 | 6.0 | 1329 | 0.1984 |
| 0.1011 | 7.0 | 1550 | 0.2254 |
| 0.0927 | 8.0 | 1772 | 0.1606 |
| 0.0735 | 9.0 | 1993 | 0.1404 |
| 0.0639 | 9.98 | 2210 | 0.1505 |
Framework versions
- Transformers 4.39.3
- Pytorch 2.1.2
- Datasets 2.18.0
- Tokenizers 0.15.2