Quantization made by Richard Erkhov.
MFANN3bv0.20.17 - bnb 4bits
- Model creator: https://huggingface.co/netcat420/
- Original model: https://huggingface.co/netcat420/MFANN3bv0.20.17/
Original model description:
library_name: transformers license: mit datasets: - netcat420/MFANN language: - en base_model: - netcat420/MFANN3bv0.20.16 pipeline_tag: text-generation
system prompt: Instruct: {instruction} Output:
good training loss of 0.33 avg on the MFANN dataset! as of 9/20/24
this model is a general model trained to generate text on a vast array of tasks and will be the last pre release in the 3b branch before the first major release!