Files
DeepSeek-R1-Distill-Merge-Q…/mergekit_config.yml

17 lines
320 B
YAML
Raw Permalink Normal View History

models:
- model: Qwen/Qwen2.5-Math-1.5B-Instruct
parameters:
weight: 0.5
- model: DeepSeek-AI/DeepSeek-R1-Distill-Qwen-1.5B
parameters:
weight: 0.5
merge_method: dare_ties
base_model: Qwen/Qwen2.5-Math-1.5B-Instruct
parameters:
density: 0.53
int8_mask: true
normalize: true
dtype: bfloat16