Quantization made by Richard Erkhov.
Jamba-tiny-random - bnb 8bits
- Model creator: https://huggingface.co/ai21labs/
- Original model: https://huggingface.co/ai21labs/Jamba-tiny-random/
Original model description:
license: apache-2.0
This is a tiny, dummy version of Jamba, used for debugging and experimentation over the Jamba architecture.
It has 128M parameters (instead of 52B), and is initialized with random weights and did not undergo any training.