10 lines
242 B
YAML
10 lines
242 B
YAML
# TIES merge: Judge as base, inject sparse GRPO knowledge
|
|
merge_method: ties
|
|
base_model: ./merged_models/llm-judge-merged-fixed
|
|
models:
|
|
- model: ./merged_models/grpo-cabs
|
|
parameters:
|
|
density: 0.5
|
|
weight: 0.4
|
|
dtype: bfloat16
|