Downloads · 30 days
0
wikitext2-bpe-tokenizer
Rogarcia18/wikitext2-bpe-tokenizer
wikitext2-bpe-tokenizer is a machine learning model from Rogarcia18. Use it for the machine learning task on the model card, and read the license before you ship it in a product. The card lists the license as apache-2.0.
A Byte Pair Encoding (BPE) tokenizer trained on the WikiText-2 dataset.
Access
Public
Updated Sep 19, 2025
Repo size
—
Likes
0
Public
Repo makeup
Click a slice to open those files.
.json2.1 MB · 100%
At a glance
- License
- apache-2.0
- Access
- Public
- Created
- Sep 19, 2025
- Updated
- Sep 19, 2025
- SHA
- 7cd0c069
Model card
From the Hugging Face model README
WikiText-2 BPE Tokenizer
A Byte Pair Encoding (BPE) tokenizer trained on the WikiText-2 dataset.
Model Details
- Vocabulary Size: 30,000 tokens
- Training Data: WikiText-2 (Salesforce/wikitext)
- Special Tokens: [PAD], [UNK], [CLS], [SEP], [MASK]
- Compression Ratio: ~6.4 characters per token
Usage
from transformers import AutoTokenizer
tokenizer = AutoTokenizer.from_pretrained("Rogarcia18/wikitext2-bpe-tokenizer")
Training Details
- Dataset: WikiText-2 (wikitext-2-v1)
- Preprocessing: Deduplication, <unk> removal, whitespace normalization, remove samples cases with less than 10 characters
- Architecture: BPE with HuggingFace tokenizers library
- License
- apache-2.0
- Created
- Sep 19, 2025
- Updated
- Sep 19, 2025