File size: 5,276 Bytes
3ef5099
3c49da8
 
 
 
 
3ef5099
 
 
 
 
 
3c49da8
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
3ef5099
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
12a2a40
6cc1389
3c49da8
 
 
 
 
 
 
 
 
 
 
 
 
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
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
---
license: llama3
library_name: transformers
tags:
- mergekit
- merge
base_model:
- nbeerbower/llama3.1-gutenberg-8B
- akjindal53244/Llama-3.1-Storm-8B
- NousResearch/Meta-Llama-3.1-8B
- nbeerbower/llama3.1-airoboros3.2-QDT-8B
- Sao10K/Llama-3.1-8B-Stheno-v3.4
model-index:
- name: Llama-3.1-8B-Ultra-Instruct
  results:
  - task:
      type: text-generation
      name: Text Generation
    dataset:
      name: IFEval (0-Shot)
      type: HuggingFaceH4/ifeval
      args:
        num_few_shot: 0
    metrics:
    - type: inst_level_strict_acc and prompt_level_strict_acc
      value: 80.81
      name: strict accuracy
    source:
      url: https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=Dampfinchen/Llama-3.1-8B-Ultra-Instruct
      name: Open LLM Leaderboard
  - task:
      type: text-generation
      name: Text Generation
    dataset:
      name: BBH (3-Shot)
      type: BBH
      args:
        num_few_shot: 3
    metrics:
    - type: acc_norm
      value: 32.49
      name: normalized accuracy
    source:
      url: https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=Dampfinchen/Llama-3.1-8B-Ultra-Instruct
      name: Open LLM Leaderboard
  - task:
      type: text-generation
      name: Text Generation
    dataset:
      name: MATH Lvl 5 (4-Shot)
      type: hendrycks/competition_math
      args:
        num_few_shot: 4
    metrics:
    - type: exact_match
      value: 14.95
      name: exact match
    source:
      url: https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=Dampfinchen/Llama-3.1-8B-Ultra-Instruct
      name: Open LLM Leaderboard
  - task:
      type: text-generation
      name: Text Generation
    dataset:
      name: GPQA (0-shot)
      type: Idavidrein/gpqa
      args:
        num_few_shot: 0
    metrics:
    - type: acc_norm
      value: 5.59
      name: acc_norm
    source:
      url: https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=Dampfinchen/Llama-3.1-8B-Ultra-Instruct
      name: Open LLM Leaderboard
  - task:
      type: text-generation
      name: Text Generation
    dataset:
      name: MuSR (0-shot)
      type: TAUR-Lab/MuSR
      args:
        num_few_shot: 0
    metrics:
    - type: acc_norm
      value: 8.61
      name: acc_norm
    source:
      url: https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=Dampfinchen/Llama-3.1-8B-Ultra-Instruct
      name: Open LLM Leaderboard
  - task:
      type: text-generation
      name: Text Generation
    dataset:
      name: MMLU-PRO (5-shot)
      type: TIGER-Lab/MMLU-Pro
      config: main
      split: test
      args:
        num_few_shot: 5
    metrics:
    - type: acc
      value: 31.4
      name: accuracy
    source:
      url: https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=Dampfinchen/Llama-3.1-8B-Ultra-Instruct
      name: Open LLM Leaderboard
---
# merge

This is a merge of pre-trained language models created using [mergekit](https://github.com/cg123/mergekit).

## Merge Details
### Merge Method

This model was merged using the [DARE](https://arxiv.org/abs/2311.03099) [TIES](https://arxiv.org/abs/2306.01708) merge method using [NousResearch/Meta-Llama-3.1-8B](https://huggingface.co/NousResearch/Meta-Llama-3.1-8B) as a base.

### Models Merged

The following models were included in the merge:
* [nbeerbower/llama3.1-gutenberg-8B](https://huggingface.co/nbeerbower/llama3.1-gutenberg-8B)
* [akjindal53244/Llama-3.1-Storm-8B](https://huggingface.co/akjindal53244/Llama-3.1-Storm-8B)
* [nbeerbower/llama3.1-airoboros3.2-QDT-8B](https://huggingface.co/nbeerbower/llama3.1-airoboros3.2-QDT-8B)
* [Sao10K/Llama-3.1-8B-Stheno-v3.4](https://huggingface.co/Sao10K/Llama-3.1-8B-Stheno-v3.4)

### Configuration

The following YAML configuration was used to produce this model:

```yaml
models:
  - model: Sao10K/Llama-3.1-8B-Stheno-v3.4
    parameters:
      weight: 0.2
      density: 0.5
  - model: akjindal53244/Llama-3.1-Storm-8B
    parameters:
      weight: 0.5
      density: 0.5
  - model: nbeerbower/llama3.1-gutenberg-8B
    parameters:
      weight: 0.3
      density: 0.5
  - model: nbeerbower/llama3.1-airoboros3.2-QDT-8B
    parameters:
      weight: 0.2
      density: 0.5
merge_method: dare_ties
base_model: NousResearch/Meta-Llama-3.1-8B
dtype: bfloat16
name: Llama-3.1-8B-Ultra-Instruct
```

Use Llama 3 Instruct prompt template. Use with caution, I'm not responsible for what you do with it. All credits and thanks go to the creators of the fine tunes I've merged. In my own tests and on HF Eval it performs very well for a 8B model and I can recommend it. High quality quants by Bartowski: https://huggingface.co/bartowski/Llama-3.1-8B-Ultra-Instruct-GGUF
# [Open LLM Leaderboard Evaluation Results](https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard)
Detailed results can be found [here](https://huggingface.co/datasets/open-llm-leaderboard/details_Dampfinchen__Llama-3.1-8B-Ultra-Instruct)

|      Metric       |Value|
|-------------------|----:|
|Avg.               |28.98|
|IFEval (0-Shot)    |80.81|
|BBH (3-Shot)       |32.49|
|MATH Lvl 5 (4-Shot)|14.95|
|GPQA (0-shot)      | 5.59|
|MuSR (0-shot)      | 8.61|
|MMLU-PRO (5-shot)  |31.40|