---
tags:
- generated_from_trainer
model-index:
- name: GPT2-705M
  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. -->

# GPT2-705M

This model is a fine-tuned version of [](https://huggingface.co/) on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 3.6046

## 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: 0.00025
- train_batch_size: 16
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 8
- total_train_batch_size: 128
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- lr_scheduler_warmup_steps: 300
- num_epochs: 20
- mixed_precision_training: Native AMP

### Training results

| Training Loss | Epoch | Step | Validation Loss |
|:-------------:|:-----:|:----:|:---------------:|
| 8.0336        | 1.0   | 3    | 7.3770          |
| 6.2535        | 2.0   | 6    | 6.3128          |
| 5.6213        | 3.0   | 9    | 5.6716          |
| 4.8242        | 4.0   | 12   | 5.1521          |
| 4.6266        | 5.0   | 15   | 4.9789          |
| 4.4097        | 6.0   | 18   | 4.7306          |
| 4.0358        | 7.0   | 21   | 4.5332          |
| 4.0027        | 8.0   | 24   | 4.4014          |
| 3.8638        | 9.0   | 27   | 4.1175          |
| 3.5414        | 10.0  | 30   | 4.0355          |
| 3.4701        | 11.0  | 33   | 3.8834          |
| 3.4822        | 12.0  | 36   | 3.8336          |
| 3.0602        | 13.0  | 39   | 3.7213          |
| 3.1109        | 14.0  | 42   | 3.7379          |
| 2.9087        | 15.0  | 45   | 3.7389          |
| 2.7124        | 16.0  | 48   | 3.6220          |
| 2.5867        | 17.0  | 51   | 3.7192          |
| 2.4577        | 18.0  | 54   | 3.5953          |
| 2.279         | 19.0  | 57   | 3.7648          |
| 2.3218        | 20.0  | 60   | 3.6046          |


### Framework versions

- Transformers 4.39.1
- Pytorch 2.1.2+cu121
- Datasets 2.16.1
- Tokenizers 0.15.0