This repository hosts GGUF-Imatrix quantizations for
ChaoticNeutrals/BuRP_7B.
It stands for
Importance Matrix, a technique used to improve the quality of quantized models.
The
Imatrix is calculated based on calibration data, and it helps determine the importance of different model activations during the quantization process.
The idea is to preserve the most important information during quantization, which can help reduce the loss of model performance, especially when the calibration data is diverse.
[1] [2]1 quantization_options = [
2 "Q4_K_M", "IQ4_XS", "Q5_K_M", "Q5_K_S", "Q6_K",
3 "Q8_0", "IQ3_M", "IQ3_S", "IQ3_XXS"
4 ]
If you want anything that's not here or another model, feel free to request.
For imatrix data generation, kalomaze's
groups_merged.txt with added roleplay chats was used, you can find it
here.
So you want a model that can do it all? You've been dying to RP with a superintelligence who never refuses your advances while sticking to your strange and oddly specific dialogue format?
Well, look no further because BuRP is the model you need.
1slices:
2 - sources:
3 - model: ErisLaylaSLERP
4 layer_range: [0, 32]
5 - model: ParadigmInfinitySLERP
6 layer_range: [0, 32]
7merge_method: slerp
8base_model: ParadigmInfinitySLERP
9parameters:
10 t:
11 - filter: self_attn
12 value: [0, 0.5, 0.3, 0.7, 1]
13 - filter: mlp
14 value: [1, 0.5, 0.7, 0.3, 0]
15 - value: 0.5
16dtype: bfloat16
