A newer version of this model is available: qikp/pika-4
pika
🎉 You are looking at pika 3, which uses the wordmix dataset!
pika is a simple and public domain-like tokenizer.
Special Tokens
- End-of-Sequence token:
[EOS](ID 0) - Padding token:
[PAD](ID 1)
Training
pika was trained on qikp/wordmix.
Limitations
Some uncommon special tokens aren't present, you'll have to add them manually if needed.
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support