File size: 399 Bytes
20b616f
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
models:
  - model: NousResearch/Llama-2-7b-hf
    #no parameters necessary for base model
  - model: arcee-ai/Patent-Instruct-7b
    parameters:
      density: 0.5
      weight: 0.5
  - model: chargoddard/internlm2-7b-llama
    parameters:
      density: 0.5
      weight: 0.5

merge_method: ties
base_model: NousResearch/Llama-2-7b-hf
parameters:
  normalize: false
  int8_mask: true
dtype: float16