Instructions to use SnurfyAI/is-even with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use SnurfyAI/is-even with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-classification", model="SnurfyAI/is-even")# Load model directly from transformers import AutoTokenizer, AutoModelForSequenceClassification tokenizer = AutoTokenizer.from_pretrained("SnurfyAI/is-even") model = AutoModelForSequenceClassification.from_pretrained("SnurfyAI/is-even") - Notebooks
- Google Colab
- Kaggle
File size: 1,541 Bytes
fff93ed de0420d fff93ed 9c9222d fff93ed a66d68a fff93ed c6cee94 fff93ed c6cee94 fff93ed c6cee94 fff93ed c6cee94 fff93ed c6cee94 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 | ---
library_name: transformers
pipeline_tag: text-classification
tags:
- transformers
- pytorch
- text-classification
- synthetic-data
- parity
license: mit
---
# Is Even
`Is Even` is a tiny decoder-only transformer trained on chat-formatted synthetic data from SnurfyAI that predicts whether a decimal input is even.
This repository contains the published model weights and metadata for the released checkpoint.
## What The Model Does
- Input: a decimal numeral rendered inside a fixed ChatML prompt
- Output: `true` for even numbers and `false` for odd numbers
- Tokenization: strict digit-level tokenization with a tiny fixed vocabulary
The task is mathematically determined by the final decimal digit, but the model was trained as a small transformer classifier on synthetic examples.
## Prompt Format
```text
<|im_start|>user
123456<|im_end|>
<|im_start|>assistant
```
The expected completion label is `true` or `false`.
## Training Code
You can train the model yourself using the official SnurfyAI training codebase:
- [SnurfyAI Is Even Core](https://github.com/SnurfyAI/is-even)
The repository contains the code needed to generate the dataset, train the model, and benchmark saved checkpoints.
## Research Paper
- [A Microscopic Study of Decimal Parity in a Tiny Causal Transformer](https://ai.snurfy.net/research/is-even)
The accompanying paper analyzes the model’s generalization behavior, attention structure, and causal dependence on the final digit.
|