---
license: mit
language:
- spa  # ISO 639-3 code or "und" if not identifiable
tags:
- tokenizer
- bpe
- flexitok
- fineweb2
---

# Byte-Level BPE Tokenizer: spa_Latn (16K)

A **Byte-Level BPE** tokenizer trained on **spa_Latn** data from Fineweb-2-HQ.

## Training Details

| Parameter | Value |
|-----------|-------|
| Algorithm | Byte-Level BPE |
| Language | `spa_Latn` |
| Target Vocab Size | 16,000 |
| Final Vocab Size | 16,955 |
| Pre-tokenizer | custom:spa_Latn |
| Number handling | ltr_3digit |
| Contraction handling | True |
| Normalizer | NFC |
| Special Tokens | `<s>`, `</s>`, `<pad>`, `<unk>` |
| Training Shards | 2 |

## Usage

```python
from transformers import AutoTokenizer

tokenizer = AutoTokenizer.from_pretrained("flexitok/bpe_ltr_spa_Latn_16000_v2")
tokens = tokenizer.encode("Hello, world!")
```

## Files

- `tokenizer.json` — Full HuggingFace tokenizer
- `vocab.json` — Vocabulary mapping
- `merges.txt` — BPE merge rules

## Sample Encoding
| Text | Tokens | Token IDs |
|------|--------|-----------|
| `Hello, world! 12345 This is a test. こんにちは` | `H, ello, ,, Ġw, orld, !, Ġ, 123, 45, ĠTh, is, Ġis, Ġa, Ġtest, ., Ġ, ãģ, ĵ, ã, Ĥ` | `42, 2841, 14, 930, 6624, 3, 223, 16808, 4579, 3182, 353, 1625, 269, 6477, 16, 223, 10858, 244, 162, 227` |