Resolving Interference When Merging Models
Paper • 2306.01708 • Published • 17
This is a merge of pre-trained language models created using mergekit.
This model was merged using the TIES merge method using Undi95/Meta-Llama-3-8B-hf as a base.
The following models were included in the merge:
The following YAML configuration was used to produce this model:
models:
- model: nothingiisreal/L3-8B-Stheno-Horny-v3.3-32K
parameters:
density: 0.5
weight: 0.5
- model: Sao10K/L3-8B-Stheno-v3.2
parameters:
density: 0.5
weight: 0.5
merge_method: ties
base_model: Undi95/Meta-Llama-3-8B-hf
parameters:
normalize: false
int8_mask: true
dtype: float16