metadata
license: mit
base_model: gpt2
tags:
- generated_from_trainer
model-index:
- name: distily_bench_gpt2_activation_loss_b
results: []
distily_bench_gpt2_activation_loss_b
This model is a fine-tuned version of gpt2 on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 1.1156
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: 4e-05
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: constant
- num_epochs: 1.0
Training results
Training Loss | Epoch | Step | Validation Loss |
---|---|---|---|
No log | 0 | 0 | 6.0052 |
1.8938 | 0.0808 | 1000 | 1.9915 |
1.7378 | 0.1616 | 2000 | 1.8007 |
1.7337 | 0.2424 | 3000 | 1.6868 |
1.5761 | 0.3232 | 4000 | 1.5975 |
1.5284 | 0.4040 | 5000 | 1.5055 |
1.3739 | 0.4848 | 6000 | 1.4362 |
1.4322 | 0.5657 | 7000 | 1.3724 |
1.3508 | 0.6465 | 8000 | 1.3171 |
1.3875 | 0.7273 | 9000 | 1.2684 |
1.3039 | 0.8081 | 10000 | 1.2160 |
1.2591 | 0.8889 | 11000 | 1.1656 |
1.2144 | 0.9697 | 12000 | 1.1263 |
Framework versions
- Transformers 4.44.0
- Pytorch 2.3.0
- Datasets 2.21.0
- Tokenizers 0.19.1