This is a merge of pre-trained language models created using
mergekit.
This model was merged using the rescaled_sample merge method using
mistralai/Mistral-7B-v0.1 as a base.
1models:
2 - model: Nexusflow/Starling-LM-7B-beta # Reasoning | OpenChat
3 parameters:
4 weight: 0.6
5 density: 0.7
6 - model: jan-hq/supermario-slerp-v3 # Reasoning | ChatML
7 parameters:
8 weight: 0.3
9 density: 0.5
10 - model: Endevor/InfinityRP-v1-7B # Roleplay | Alpaca
11 parameters:
12 weight: 0.3
13 density: 0.5
14 - model: NeverSleep/Noromaid-7B-0.4-DPO # Roleplay | ChatML
15 parameters:
16 weight: 0.2
17 density: 0.4
18merge_method: rescaled_sample
19base_model: mistralai/Mistral-7B-v0.1
20parameters:
21 int8_mask: true
22dtype: bfloat16
23
