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 openai-community/gpt2-xl as a base.
The following models were included in the merge:
The following YAML configuration was used to produce this model:
models:
- model: openai-community/gpt2-xl
parameters:
density: 0.5
weight: 0.5
- model: openai-community/gpt2-xl
parameters:
density: 0.5
weight: 0.5
merge_method: ties
base_model: openai-community/gpt2-xl
parameters:
normalize: false
int8_mask: true
dtype: float16
Base model
openai-community/gpt2-xl