metadata
license: mit
base_model: gpt2
tags:
- generated_from_trainer
model-index:
- name: constant_with_warmup_2000_9e-4_16b_w0.08
results: []
constant_with_warmup_2000_9e-4_16b_w0.08
This model is a fine-tuned version of gpt2 on the None dataset. It achieves the following results on the evaluation set:
- Loss: 2.8710
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: 0.0009
- train_batch_size: 16
- eval_batch_size: 16
- seed: 42
- gradient_accumulation_steps: 16
- total_train_batch_size: 256
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: constant_with_warmup
- lr_scheduler_warmup_steps: 250
- training_steps: 2000
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 6.293 | 0.7930 | 250 | 4.7425 |
| 4.0889 | 1.5860 | 500 | 3.5180 |
| 3.2871 | 2.3791 | 750 | 3.1743 |
| 3.0032 | 3.1721 | 1000 | 3.0286 |
| 2.8278 | 3.9651 | 1250 | 2.9422 |
| 2.6539 | 4.7581 | 1500 | 2.9041 |
| 2.5482 | 5.5511 | 1750 | 2.8808 |
| 2.4602 | 6.3442 | 2000 | 2.8710 |
Framework versions
- Transformers 4.40.1
- Pytorch 2.3.0+cu121
- Datasets 2.19.0
- Tokenizers 0.19.1