Experimental model, trained using config and
Transformers/Axolotl forks provided by
Doctor-ShotgunModel was fine-tuned from
Mixtral-8x7B-v0.1 with airoboros-3.2 dataset, for 4 epochs, ChatML prompt format at 8K context length.
Additionally, model was then merged with
Mixtral-8x7B-Instruct-v0.1:
This is a merge of pre-trained language models created using
mergekit.
This model was merged using the
linear merge method.
1models:
2 - model: mistralai/Mixtral-8x7B-Instruct-v0.1
3 parameters:
4 weight: 0.5
5 - model: LoneStriker/Air-Striker-Mixtral-8x7B-ZLoss
6 parameters:
7 weight: 0.5
8merge_method: linear
9dtype: bfloat16
