An experiment with gradient merges using the following script, with Chronos as its primary model, augmented by Hermes and Wizard-Vicuna Uncensored.
Chronos is a wonderful model, though doesn't feel very smart. Hermes and WizardLM have been merged gradually, primarily in the higher layers (10+) in an attempt to rectify some of this behaviour without affecting Chronos' lengthy replies. I'd say the end product is about 60% Chronos, with 20% Hermes and 20% Wizard added in gradually increasing amounts.
A 4_K_M quant has been included for convenience sake. Happy experimenting!
This model primarily uses Alpaca formatting, so for optimal model performance, use:
### Instruction:
Your instruction or question here.
### Response: