della
della copied to clipboard
DELLA-Merging: Reducing Interference in Model Merging through Magnitude-Based Sampling
Results
2
della issues
Sort by
recently updated
recently updated
newest added
models: - model: models/vicuna-7b-v1.5-16k parameters: weight: 1.0 - model: models/vicuna_7b_A parameters: weight: 1.0 base_model: models/vicuna-7b-v1.5-16k merge_method: della parameters: normalize: true int8_mask: true density: 0.7 lambda: 1.1 epsilon: 0.2 dtype: float16...
What are lambda and epsilon in the YAML config?