Importance Matrix, at least based on my testing, has shown to improve the output and performance of "IQ"-type quantizations, where the compression becomes quite heavy. The Imatrix performs a calibration, using a provided dataset. Testing has shown that semi-randomized data can help perserve more important segments as the compression is applied.

Related discussions in Github: [1] [2]

The imatrix.txt file that I used contains general, semi-random data, with some custom kink.

The concept behind this merge was to use the improved reasoning of of Cerebral-Infinity-7B, and merge it with the improved originality of Infinite-Laymons-7B.

I think the experiment worked, and so far I am happy with the results.

This model is intended for fictional storytelling and role-playing, with a focus on more original conversations and less alignment.

This is a merge of pre-trained language models created using mergekit.

This model was merged using the passthrough merge method.

The following models were included in the merge:

The following YAML configuration was used to produce this model: