RichardErkhov/chihoonlee10_-_T3Q-MSlerp-13B-gguf

Quantization made by Richard Erkhov.

Github

Discord

Request more models

T3Q-MSlerp-13B - GGUF

• Model creator: https://huggingface.co/chihoonlee10/
• Original model: https://huggingface.co/chihoonlee10/T3Q-MSlerp-13B/

Name	Quant method	Size
T3Q-MSlerp-13B.Q2_K.gguf	Q2_K	4.43GB
T3Q-MSlerp-13B.IQ3_XS.gguf	IQ3_XS	4.95GB
T3Q-MSlerp-13B.IQ3_S.gguf	IQ3_S	5.22GB
T3Q-MSlerp-13B.Q3_K_S.gguf	Q3_K_S	5.2GB
T3Q-MSlerp-13B.IQ3_M.gguf	IQ3_M	5.35GB
T3Q-MSlerp-13B.Q3_K.gguf	Q3_K	5.78GB
T3Q-MSlerp-13B.Q3_K_M.gguf	Q3_K_M	5.78GB
T3Q-MSlerp-13B.Q3_K_L.gguf	Q3_K_L	6.27GB
T3Q-MSlerp-13B.IQ4_XS.gguf	IQ4_XS	6.5GB
T3Q-MSlerp-13B.Q4_0.gguf	Q4_0	6.78GB
T3Q-MSlerp-13B.IQ4_NL.gguf	IQ4_NL	6.85GB
T3Q-MSlerp-13B.Q4_K_S.gguf	Q4_K_S	6.84GB
T3Q-MSlerp-13B.Q4_K.gguf	Q4_K	7.25GB
T3Q-MSlerp-13B.Q4_K_M.gguf	Q4_K_M	7.25GB
T3Q-MSlerp-13B.Q4_1.gguf	Q4_1	7.52GB
T3Q-MSlerp-13B.Q5_0.gguf	Q5_0	8.26GB
T3Q-MSlerp-13B.Q5_K_S.gguf	Q5_K_S	8.26GB
T3Q-MSlerp-13B.Q5_K.gguf	Q5_K	8.51GB
T3Q-MSlerp-13B.Q5_K_M.gguf	Q5_K_M	8.51GB
T3Q-MSlerp-13B.Q5_1.gguf	Q5_1	9.01GB
T3Q-MSlerp-13B.Q6_K.gguf	Q6_K	9.84GB
T3Q-MSlerp-13B.Q8_0.gguf	Q8_0	12.75GB

Original model description:

license: apache-2.0

T3Q-MSlerp-13B

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

• zhengr/MixTAO-7Bx2-MoE-Instruct-v7.0
• yunconglong/13B_MATH_DPO

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