marlosb/smol-smoltalk-pt

---
license: apache-2.0
language:
- en
tags:
- synthetic
pretty_name: Smol-SmolTalk
size_categories:
- 100K<n<1M
dataset_info:
features:
- name: messages
list:
- name: content
dtype: string
- name: role
dtype: string
- name: source
dtype: string
splits:
- name: train
num_bytes: 1789024390
num_examples: 457307
- name: test
num_bytes: 93616814
num_examples: 24035
download_size: 949202245
dataset_size: 1882641204
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*
---
# marlosb/smol-smoltalk-pt

This dataset is a Portuguese translation of the original **Smol-SmolTalk** dataset from HuggingFaceTB.

## Original Dataset
- **Hugging Face**: [HuggingFaceTB/smol-smoltalk](https://huggingface.co/datasets/HuggingFaceTB/smol-smoltalk)

## Dataset Summary
This is a subset of [SmolTalk](https://huggingface.co/datasets/HuggingFaceTB/smoltalk/) dataset adapted for smol models with less than 1B parameters.

Compared to SmolTalk:
- The conversations from Smol-Magpie-Ultra are shorter in this dataset
- We include less task-specific data compared to SmolTalk (e.g., no function calling and less rewriting and summarization examples) since these smaller models have limited capacity
- We don't include any advanced math datasets

This translated version preserves the exact structure, splits, and fields. Conversations were translated to Portuguese while maintaining the original format.

## License
**Apache 2.0** (same as the original)

```python
from datasets import load_dataset

ds = load_dataset("marlosb/smol-smoltalk-pt", split="train")