metadata
license: apache-2.0
base_model: distilgpt2
tags:
- generated_from_trainer
model-index:
- name: judge_JuDe
results: []
judge_JuDe
This model is a fine-tuned version of distilgpt2 on the None dataset. It achieves the following results on the evaluation set:
- Loss: 2.4628
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: 2e-05
- train_batch_size: 16
- eval_batch_size: 16
- 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 |
|---|---|---|---|
| 2.9881 | 0.25 | 2057 | 2.7543 |
| 2.803 | 0.5 | 4114 | 2.6652 |
| 2.7298 | 0.75 | 6171 | 2.6124 |
| 2.687 | 1.0 | 8228 | 2.5772 |
| 2.6374 | 1.25 | 10285 | 2.5535 |
| 2.6161 | 1.5 | 12342 | 2.5332 |
| 2.598 | 1.75 | 14399 | 2.5171 |
| 2.5773 | 2.0 | 16456 | 2.5050 |
| 2.5578 | 2.25 | 18513 | 2.4943 |
| 2.5468 | 2.5 | 20570 | 2.4868 |
| 2.5385 | 2.75 | 22627 | 2.4783 |
| 2.5322 | 3.0 | 24684 | 2.4712 |
| 2.5182 | 3.25 | 26741 | 2.4697 |
| 2.5188 | 3.5 | 28798 | 2.4657 |
| 2.513 | 3.75 | 30855 | 2.4630 |
| 2.5123 | 4.0 | 32912 | 2.4628 |
Framework versions
- Transformers 4.34.1
- Pytorch 1.12.1+cu113
- Datasets 2.8.0
- Tokenizers 0.14.1