metadata
license: mit
library_name: peft
tags:
- generated_from_trainer
datasets:
- emotion
base_model: gpt2
model-index:
- name: output
results: []
output
This model is a fine-tuned version of gpt2 on the emotion dataset. It achieves the following results on the evaluation set:
- Loss: 3.8303
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: 0.001
- 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 |
|---|---|---|---|
| No log | 0.25 | 250 | 4.8185 |
| No log | 0.5 | 500 | 4.3814 |
| No log | 0.75 | 750 | 4.1230 |
| No log | 1.0 | 1000 | 4.0088 |
| No log | 1.25 | 1250 | 3.9536 |
| No log | 1.5 | 1500 | 3.9208 |
| No log | 1.75 | 1750 | 3.8946 |
| 4.644 | 2.0 | 2000 | 3.8799 |
| 4.644 | 2.25 | 2250 | 3.8651 |
| 4.644 | 2.5 | 2500 | 3.8552 |
| 4.644 | 2.75 | 2750 | 3.8464 |
| 4.644 | 3.0 | 3000 | 3.8399 |
| 4.644 | 3.25 | 3250 | 3.8364 |
| 4.644 | 3.5 | 3500 | 3.8333 |
| 4.644 | 3.75 | 3750 | 3.8311 |
| 4.0742 | 4.0 | 4000 | 3.8303 |
Framework versions
- PEFT 0.7.1
- Transformers 4.36.2
- Pytorch 2.1.2
- Datasets 2.15.0
- Tokenizers 0.15.1