File size: 2,190 Bytes
bfe59bb
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
---
base_model:
- meta-llama/Meta-Llama-3-70B-Instruct
license: llama3
language:
- en
pipeline_tag: text-generation
tags:
- merge
- frankenmerge
- 96b
---
# BigWeave v33 105b

<img src="https://cdn-uploads.huggingface.co/production/uploads/65a6db055c58475cf9e6def1/4CbbAN-X7ZWj702JrcCGH.png" width=600>

The BigWeave models aim to experimentally identify merge settings for increasing model performance. The version number merely tracks various attempts and is not a quality indicator. Only results demonstrating good performance are retained and shared.

# Prompting Format
llamav3

# Merge process
This is a self-merge of meta-llama/Meta-Llama-3-70B-Instruct. Middle layers are duplicated and various matrices are scaled according to the template by jukofyork as shown here: https://github.com/arcee-ai/mergekit/issues/198#issuecomment-2079950009

Merge configuration:
```
const_tag: &MODEL meta-llama/Meta-Llama-3-70B-Instruct

const_tag: &RESIDUAL_SCALE_FACTOR 0.5
const_tag: &QK_ATTENUATION_FACTOR 0.7071067812
const_tag: &OUT_FACTOR 0.9

scale-filter-env: &scale_filter_env
  parameters:
    scale:
      - filter: o_proj
        value: *RESIDUAL_SCALE_FACTOR
      - filter: down_proj
        value: *RESIDUAL_SCALE_FACTOR
      - filter: q_proj
        value: *QK_ATTENUATION_FACTOR
      - filter: k_proj
        value: *QK_ATTENUATION_FACTOR
      - filter: v_proj
        value: *OUT_FACTOR
      - filter: up_proj
        value: *OUT_FACTOR
      - value: 1.0

slices:
  - sources:
    - model: *MODEL
      layer_range: [0, 19]
  - sources:
    - model: *MODEL
      layer_range: [19, 20]
      <<: *scale_filter_env

  - sources:
    - model: *MODEL
      layer_range: [10, 29]
  - sources:
    - model: *MODEL
      layer_range: [29, 30]
      <<: *scale_filter_env

  - sources:
    - model: *MODEL
      layer_range: [20, 39]
  - sources:
    - model: *MODEL
      layer_range: [39, 40]
      <<: *scale_filter_env

  - sources:
    - model: *MODEL
      layer_range: [30, 49]
  - sources:
    - model: *MODEL
      layer_range: [49, 50]
      <<: *scale_filter_env

  - sources:
    - model: *MODEL
      layer_range: [40, 80]

merge_method: passthrough
dtype: float16
```