File size: 1,690 Bytes
fe97d6b
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
merge_method: task_arithmetic
base_model: meta-llama/Llama-3.3-70B-Instruct
models:
  - model: meta-llama/Llama-3.1-70B
    parameters:
      weight: -1.0
  - model: tokyotech-llm/Llama-3.1-Swallow-70B-v0.1
    parameters:
      weight: 1.0
dtype: bfloat16
name: Llama-3.3-SuperSwallow-70B-Instruct-v0.1-preset-llama33
---
merge_method: task_arithmetic
base_model: Llama-3.3-SuperSwallow-70B-Instruct-v0.1-preset-llama33
models:
  - model: meta-llama/Llama-3.1-70B-Instruct
    parameters:
      weight: -1.0
  - model: nvidia/Llama-3.1-Nemotron-70B-Instruct-HF
    parameters:
      weight: 1.0
dtype: bfloat16
name: Llama-3.3-SuperSwallow-70B-Instruct-v0.1-preset-nemollama33
---
merge_method: task_arithmetic
base_model: Llama-3.3-SuperSwallow-70B-Instruct-v0.1-preset-nemollama33
models:
  - model: meta-llama/Llama-3.1-70B
    parameters:
      weight: -0.8
  - model: allenai/Llama-3.1-Tulu-3-70B
    parameters:
      weight: 0.8
dtype: bfloat16
name: Llama-3.3-SuperSwallow-70B-Instruct-v0.1-preset-tulu
---
merge_method: task_arithmetic
base_model: Llama-3.3-SuperSwallow-70B-Instruct-v0.1-preset-nemollama33
models:
  - model: tokyotech-llm/Llama-3.1-Swallow-70B-v0.1
    parameters:
      weight: -0.8
  - model: tokyotech-llm/Llama-3.1-Swallow-70B-Instruct-v0.1
    parameters:
      weight: 0.8
dtype: bfloat16
name: Llama-3.3-SuperSwallow-70B-Instruct-v0.1-preset-swallow
---
merge_method: model_stock
base_model: Llama-3.3-SuperSwallow-70B-Instruct-v0.1-preset-nemollama33
models:
  - model: Llama-3.3-SuperSwallow-70B-Instruct-v0.1-preset-tulu
  - model: Llama-3.3-SuperSwallow-70B-Instruct-v0.1-preset-swallow
dtype: bfloat16
name: Llama-3.3-SuperSwallow-70B-Instruct-v0.1