metadata
license: mit
base_model: gpt2
tags:
- generated_from_trainer
model-index:
- name: polynomial_2000_9e-4_32b_w0.08_cased
results: []
polynomial_2000_9e-4_32b_w0.08_cased
This model is a fine-tuned version of gpt2 on the None dataset. It achieves the following results on the evaluation set:
- Loss: 3.1523
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: 32
- eval_batch_size: 32
- seed: 42
- gradient_accumulation_steps: 16
- total_train_batch_size: 512
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: polynomial
- lr_scheduler_warmup_steps: 250
- training_steps: 2000
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 6.3307 | 1.9203 | 250 | 4.7026 |
| 3.8482 | 3.8406 | 500 | 3.4062 |
| 3.0517 | 5.7609 | 750 | 3.1624 |
| 2.7001 | 7.6812 | 1000 | 3.0874 |
| 2.4363 | 9.6015 | 1250 | 3.0764 |
| 2.2164 | 11.5218 | 1500 | 3.0995 |
| 2.0174 | 13.4422 | 1750 | 3.1357 |
| 1.855 | 15.3625 | 2000 | 3.1523 |
Framework versions
- Transformers 4.40.1
- Pytorch 2.3.0+cu121
- Datasets 2.19.0
- Tokenizers 0.19.1