T3Q-MSlerp-13B / README.md
chlee10's picture
Update README.md
332cfb7 verified
|
raw
history blame contribute delete
No virus
879 Bytes
metadata
license: apache-2.0

T3Q-MSlerp-13B

T3Q-MSlerp-13B is a merge of the following models using mergekit:

Model Developers Chihoon Lee(chlee10), T3Q

  
  slices:
      - sources:
        - model: zhengr/MixTAO-7Bx2-MoE-Instruct-v7.0
          layer_range: [0, 32]
        - model: yunconglong/13B_MATH_DPO
          layer_range: [0, 32]
  
  merge_method: slerp
  base_model: zhengr/MixTAO-7Bx2-MoE-Instruct-v7.0
  
  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 rest of tensors
      
  dtype: float16