# Model card for CoNN Add Carry


### Introduction
In paper Neural Comprehension: Language Models with Compiled Neural Networks , we introduced the integration of Compiled Neural Networks (CoNN) into the framework of language models, enabling existing language models to perform symbolic operations with perfect accuracy without the need for external tools. 
In this model card, we introduce the Add Carry model, which is similar to the Transformer model and can perform carry operations on a sequence of numbers added in parallel.



### Install 

```
git clone https://github.com/WENGSYX/Neural-Comprehension
cd apex
pip install .
```

To run neural comprehension, you need to install `PyTorch`, `Transformers`, `jax`, and `tracr`.



### How to Use?

```
from NeuralComprehension.CoNN.modeling_conn import CoNNModel
from NeuralComprehension.tracr4torch import Tokenizer


model = CoNNModel.from_pretrained('WENGSYX/CoNN_Add_Carry')
tokenizer = Tokenizer(model.config.input_encoding_map, model.config.output_encoding_map,model.config.max_position_embeddings)

output = model(tokenizer('2 15 3 8 10').unsqueeze(0))
print(tokenizer.decode(output.argmax(2)))

>>> [['bos', '3', '5', '3', '9', '0']]
```