70 lines
1.9 KiB
Markdown
70 lines
1.9 KiB
Markdown
---
|
|
license: mit
|
|
tags:
|
|
- generated_from_trainer
|
|
metrics:
|
|
- accuracy
|
|
model-index:
|
|
- name: gpt2-sweep
|
|
results: []
|
|
---
|
|
|
|
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
should probably proofread and complete it, then remove this comment. -->
|
|
|
|
# gpt2-sweep
|
|
|
|
This model is a fine-tuned version of [gpt2-large](https://huggingface.co/gpt2-large) on the None dataset.
|
|
It achieves the following results on the evaluation set:
|
|
- Loss: 2.0808
|
|
- Accuracy: 0.8556
|
|
|
|
## 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: 2.294477077303931e-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: linear
|
|
- lr_scheduler_warmup_steps: 1000
|
|
- num_epochs: 2.0
|
|
|
|
### Training results
|
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|
|
|:-------------:|:-----:|:-----:|:---------------:|:--------:|
|
|
| 2.4827 | 0.19 | 1000 | 2.4565 | 0.8520 |
|
|
| 2.6468 | 0.37 | 2000 | 2.3303 | 0.8530 |
|
|
| 2.5106 | 0.56 | 3000 | 2.2487 | 0.8537 |
|
|
| 2.0732 | 0.74 | 4000 | 2.2020 | 0.8541 |
|
|
| 2.159 | 0.93 | 5000 | 2.1594 | 0.8545 |
|
|
| 1.856 | 1.12 | 6000 | 2.1518 | 0.8548 |
|
|
| 1.9138 | 1.3 | 7000 | 2.1261 | 0.8551 |
|
|
| 1.8055 | 1.49 | 8000 | 2.1126 | 0.8552 |
|
|
| 2.0385 | 1.67 | 9000 | 2.1008 | 0.8554 |
|
|
| 1.9648 | 1.86 | 10000 | 2.0858 | 0.8555 |
|
|
|
|
|
|
### Framework versions
|
|
|
|
- Transformers 4.26.0
|
|
- Pytorch 2.0.0+cu117
|
|
- Datasets 2.9.0
|
|
- Tokenizers 0.13.2
|