metadata

base_model: []
library_name: transformers
tags:
  - mergekit
  - merge
  - alpaca
  - mistral
  - not-for-all-audiences
  - nsfw
  - exl2
license: cc-by-nc-4.0

IceCocoaRP-7b-8bpw-exl2

8bpw-exl2 quant of icefog72/IceCocoaRP-7b

SillyTavern Discord thread

Rules-lorebook and settings I'm using you can find here

ko-fi

Merge Details

The best one so far for me.

4.2bpw-exl2
6.5bpw-exl2
8bpw-exl2

Merge Method

This model was merged using the TIES merge method using NeuralBeagleJaskier as a base.

Models Merged

The following models were included in the merge:

NeuralBeagleJaskier
IceBlendedCoffeeRP-7b (slerp bfloat16)
- IceCoffeeRP-7b
- IceBlendedLatteRP-7b base

Configuration

The following YAML configuration was used to produce this model:

models:
  - model: NeuralBeagleJaskier
    parameters:
      density: 0.9
      weight: 0.5
  - model: IceBlendedCoffeeRP-7b
    parameters:
      density: 0.5
      weight: 0.3
merge_method: ties
base_model: NeuralBeagleJaskier
parameters:
  normalize: true
  int8_mask: true
dtype: float16