sft_cml4 / README.md
osanseviero's picture
osanseviero HF staff
End of training
729c4da verified
---
license: mit
base_model: gpt2
tags:
- generated_from_trainer
datasets:
- ag_news
model-index:
- name: sft_cml4
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. -->
# sft_cml4
This model is a fine-tuned version of [gpt2](https://huggingface.co/gpt2) on the ag_news dataset.
It achieves the following results on the evaluation set:
- Loss: 3.3980
## 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: 0.0005
- 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: cosine
- num_epochs: 2
### Training results
| Training Loss | Epoch | Step | Validation Loss |
|:-------------:|:-----:|:----:|:---------------:|
| 3.7271 | 0.32 | 200 | 3.6065 |
| 3.346 | 0.64 | 400 | 3.4732 |
| 3.0685 | 0.96 | 600 | 3.3985 |
| 2.1435 | 1.28 | 800 | 3.4433 |
| 1.9834 | 1.6 | 1000 | 3.4203 |
| 1.8937 | 1.92 | 1200 | 3.3980 |
### Framework versions
- Transformers 4.34.1
- Pytorch 2.0.1+cu118
- Datasets 2.14.4
- Tokenizers 0.14.0