|
---
|
|
library_name: transformers
|
|
license: mit
|
|
base_model: gpt2
|
|
tags:
|
|
- Text generation
|
|
- generated_from_trainer
|
|
model-index:
|
|
- name: codeparrot-ds-2
|
|
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. -->
|
|
|
|
# codeparrot-ds-2
|
|
|
|
This model is a fine-tuned version of [gpt2](https://huggingface.co/gpt2) on an unknown dataset.
|
|
It achieves the following results on the evaluation set:
|
|
- Loss: 1.0614
|
|
|
|
## 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: 32
|
|
- eval_batch_size: 32
|
|
- seed: 42
|
|
- gradient_accumulation_steps: 8
|
|
- total_train_batch_size: 256
|
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
|
- lr_scheduler_type: cosine
|
|
- lr_scheduler_warmup_steps: 1000
|
|
- num_epochs: 1
|
|
- mixed_precision_training: Native AMP
|
|
|
|
### Training results
|
|
|
|
| Training Loss | Epoch | Step | Validation Loss |
|
|
|:-------------:|:------:|:-----:|:---------------:|
|
|
| 2.5681 | 0.0766 | 5000 | 1.7468 |
|
|
| 1.6828 | 0.1533 | 10000 | 1.5245 |
|
|
| 1.5337 | 0.2299 | 15000 | 1.4208 |
|
|
| 1.4542 | 0.3065 | 20000 | 1.3556 |
|
|
| 1.3941 | 0.3832 | 25000 | 1.3040 |
|
|
| 1.3429 | 0.4598 | 30000 | 1.2564 |
|
|
| 1.2972 | 0.5365 | 35000 | 1.2130 |
|
|
| 1.2508 | 0.6131 | 40000 | 1.1707 |
|
|
| 1.21 | 0.6897 | 45000 | 1.1323 |
|
|
| 1.1723 | 0.7664 | 50000 | 1.0998 |
|
|
| 1.1453 | 0.8430 | 55000 | 1.0765 |
|
|
| 1.1239 | 0.9196 | 60000 | 1.0645 |
|
|
| 1.1165 | 0.9963 | 65000 | 1.0614 |
|
|
|
|
|
|
### Framework versions
|
|
|
|
- Transformers 4.44.1
|
|
- Pytorch 2.4.0+cu118
|
|
- Datasets 2.21.0
|
|
- Tokenizers 0.19.1
|
|
|
