qwenreader3 / mergekit_config.yml
kunal732's picture
Upload 10 files
dc50bdb verified
slices:
- sources:
- model: Qwen/Qwen2.5-1.5B-Instruct
layer_range: [0, 3] # Covering all 28 layers (0 to 27)
- model: jinaai/ReaderLM-v2
layer_range: [0, 3] # Ensure this model has a similar layer structure
merge_method: slerp
base_model: jinaai/ReaderLM-v2
parameters:
t:
- filter: self_attn
value: [0, 0.5, 0.3, 0.7, 1]
- filter: mlp
value: [1, 0.5, 0.7, 0.3, 0]
- value: 0.5 # Fallback for any tensors not covered by the filters
dtype: bfloat16 # Or float16, depending on your preference and model dtype