pika
🎉 You are looking at pika 4, which is using FinePhrase and a more efficient vocabulary size!
pika is a simple and public domain-like tokenizer.
Special Tokens
- End-of-Sequence token:
<|endoftext|>(ID 0)
Training
pika was trained on a portion of format/explanation-1b-hq from FinePhrase.
Limitations
Extra 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