|
--- |
|
|
|
language: |
|
- hu |
|
tags: |
|
- text-generation |
|
license: cc-by-nc-4.0 |
|
widget: |
|
- text: "Elmesélek egy történetet a nyelvtechnológiáról." |
|
|
|
--- |
|
|
|
# PULI GPT-3SX |
|
|
|
For further details, see [our demo site](https://juniper.nytud.hu/demo/puli). |
|
|
|
- Hungarian GPT-NeoX model (6.7 billion parameter) |
|
- Trained with EleutherAI's GPT-NeoX [github](https://github.com/EleutherAI/gpt-neox) |
|
- Dataset: 36.3 billion words |
|
- Checkpoint: 150 000 step |
|
|
|
## Limitations |
|
|
|
- max_seq_length = 2048 |
|
|
|
|
|
## Citation |
|
If you use this model, please cite the following paper: |
|
|
|
``` |
|
@inproceedings {yang-gpt3, |
|
title = {Jönnek a nagyok! GPT-3, GPT-2 és BERT large nyelvmodellek magyar nyelvre}, |
|
booktitle = {XIX. Magyar Számítógépes Nyelvészeti Konferencia (MSZNY 2023)}, |
|
year = {2023}, |
|
publisher = {Szegedi Tudományegyetem}, |
|
address = {Szeged, Hungary}, |
|
author = {Yang, Zijian Győző}, |
|
pages = {0} |
|
} |
|
|
|
``` |
|
|
|
## Usage |
|
|
|
```python |
|
from transformers import GPTNeoXForCausalLM, GPTNeoXTokenizerFast |
|
|
|
model = GPTNeoXForCausalLM.from_pretrained("NYTK/PULI-GPT-3SX") |
|
tokenizer = GPTNeoXTokenizerFast.from_pretrained("NYTK/PULI-GPT-3SX") |
|
prompt = "Elmesélek egy történetet a nyelvtechnológiáról." |
|
input_ids = tokenizer(prompt, return_tensors="pt").input_ids |
|
|
|
gen_tokens = model.generate( |
|
input_ids, |
|
do_sample=True, |
|
temperature=0.9, |
|
max_length=100, |
|
) |
|
|
|
gen_text = tokenizer.batch_decode(gen_tokens)[0] |
|
print(gen_text) |
|
``` |