File size: 7,074 Bytes
bf61d89
 
 
4d78f40
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
bf61d89
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
25701f3
d2fc369
25701f3
d2fc369
25701f3
d2fc369
 
 
 
25701f3
 
 
bf61d89
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
4d78f40
 
 
 
 
 
 
 
 
 
 
 
 
 
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
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
---
language:
- en
license: apache-2.0
model-index:
- name: Mistral-7B-Instruct-v0.2-DARE
  results:
  - task:
      type: text-generation
      name: Text Generation
    dataset:
      name: AI2 Reasoning Challenge (25-Shot)
      type: ai2_arc
      config: ARC-Challenge
      split: test
      args:
        num_few_shot: 25
    metrics:
    - type: acc_norm
      value: 61.95
      name: normalized accuracy
    source:
      url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=janhq/Mistral-7B-Instruct-v0.2-DARE
      name: Open LLM Leaderboard
  - task:
      type: text-generation
      name: Text Generation
    dataset:
      name: HellaSwag (10-Shot)
      type: hellaswag
      split: validation
      args:
        num_few_shot: 10
    metrics:
    - type: acc_norm
      value: 75.62
      name: normalized accuracy
    source:
      url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=janhq/Mistral-7B-Instruct-v0.2-DARE
      name: Open LLM Leaderboard
  - task:
      type: text-generation
      name: Text Generation
    dataset:
      name: MMLU (5-Shot)
      type: cais/mmlu
      config: all
      split: test
      args:
        num_few_shot: 5
    metrics:
    - type: acc
      value: 49.99
      name: accuracy
    source:
      url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=janhq/Mistral-7B-Instruct-v0.2-DARE
      name: Open LLM Leaderboard
  - task:
      type: text-generation
      name: Text Generation
    dataset:
      name: TruthfulQA (0-shot)
      type: truthful_qa
      config: multiple_choice
      split: validation
      args:
        num_few_shot: 0
    metrics:
    - type: mc2
      value: 54.36
    source:
      url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=janhq/Mistral-7B-Instruct-v0.2-DARE
      name: Open LLM Leaderboard
  - task:
      type: text-generation
      name: Text Generation
    dataset:
      name: Winogrande (5-shot)
      type: winogrande
      config: winogrande_xl
      split: validation
      args:
        num_few_shot: 5
    metrics:
    - type: acc
      value: 74.98
      name: accuracy
    source:
      url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=janhq/Mistral-7B-Instruct-v0.2-DARE
      name: Open LLM Leaderboard
  - task:
      type: text-generation
      name: Text Generation
    dataset:
      name: GSM8k (5-shot)
      type: gsm8k
      config: main
      split: test
      args:
        num_few_shot: 5
    metrics:
    - type: acc
      value: 18.12
      name: accuracy
    source:
      url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=janhq/Mistral-7B-Instruct-v0.2-DARE
      name: Open LLM Leaderboard
---
<!-- header start -->
<!-- 200823 -->
<div style="width: auto; margin-left: auto; margin-right: auto">
<img src="https://github.com/janhq/jan/assets/89722390/35daac7d-b895-487c-a6ac-6663daaad78e" alt="Jan banner" style="width: 100%; min-width: 400px; display: block; margin: auto;">
</div>

<p align="center">
    <a
 href="https://jan.ai/">Jan</a> 
    - <a href="https://discord.gg/AsJ8krTT3N">Discord</a>
</p>
<!-- header end -->

# Model Description
This model uses the `DARE` method to merge [Mistral-7B-Instruct-v0.2](https://huggingface.co/mistralai/Mistral-7B-Instruct-v0.2) with 3 leading models in 12th Dec on [OpenLLM Leaderboard](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard):
1. [OpenHermes-2.5-neural-chat-v3-3-Slerp](https://huggingface.co/Weyaxi/OpenHermes-2.5-neural-chat-v3-3-Slerp)
2. [MetaMath-Cybertron-Starling](https://huggingface.co/Q-bert/MetaMath-Cybertron-Starling)
3. [v1olet_marcoroni-go-bruins-merge-7B](https://huggingface.co/v1olet/v1olet_marcoroni-go-bruins-merge-7B)

- base model: [Mistral-7B-Instruct-v0.2](https://huggingface.co/mistralai/Mistral-7B-Instruct-v0.2)

The yaml config file for this model is here:

```yaml
base_model: mistralai/Mistral-7B-Instruct-v0.2
dtype: bfloat16
merge_method: dare_ties
models:
- model: mistralai/Mistral-7B-Instruct-v0.2
- model: Weyaxi/OpenHermes-2.5-neural-chat-v3-3-Slerp
  parameters:
    density: 0.8
    weight: 0.4
- model: Q-bert/MetaMath-Cybertron-Starling
  parameters:
    density: 0.8
    weight: 0.3
- model: v1olet/v1olet_marcoroni-go-bruins-merge-7B
  parameters:
    density: 0.8
    weight: 0.3
parameters:
  int8_mask: true
```

# Prompt template:

- **ChatML**

```
<|im_start|>system
{system_message}<|im_end|>
<|im_start|>user
{prompt}<|im_end|>
<|im_start|>assistant
```
- **Alpaca**
```
{system_message}

### Instruction:
{prompt}

### Response:
```

# Run this model
You can run this model using [Jan Desktop](https://jan.ai/) on Mac, Windows, or Linux.

Jan is an open source, ChatGPT alternative that is:

- ๐Ÿ’ป  **100% offline on your machine**: Your conversations remain confidential, and visible only to you.
- ๐Ÿ—‚๏ธ **An Open File Format**: Conversations and model settings stay on your computer and can be exported or deleted at any time.
- ๐ŸŒ **OpenAI Compatible**: Local server on port `1337` with OpenAI compatible endpoints
- ๐ŸŒ **Open Source & Free**: We build in public; check out our [Github](https://github.com/janhq)

![image/png](https://cdn-uploads.huggingface.co/production/uploads/65713d70f56f9538679e5a56/r7VmEBLGXpPLTu2MImM7S.png)

# About Jan
Jan believes in the need for an open-source AI ecosystem and is building the infra and tooling to allow open-source AIs to compete on a level playing field with proprietary ones.

Jan's long-term vision is to build a cognitive framework for future robots, who are practical, useful assistants for humans and businesses in everyday life.

# Jan Model Merger
This is a test project for merging models.

# Open LLM Leaderboard Evaluation Results

Detailed results can be found here.

| Metric                | Value                     |
|-----------------------|---------------------------|
| Avg.                  | ?|
| ARC (25-shot)         | ?          |
| HellaSwag (10-shot)   | ?   |
| MMLU (5-shot)         | ?|
| TruthfulQA (0-shot)   | ? |
| Winogrande (5-shot)   | ?  |
| GSM8K (5-shot)        | ?        |

# Acknowlegement
- [mergekit](https://github.com/cg123/mergekit)
- [DARE](https://github.com/yule-BUAA/MergeLM/blob/main/README.md)
- [SLERP](https://github.com/Digitous/LLM-SLERP-Merge)
- [lm-evaluation-harness](https://github.com/EleutherAI/lm-evaluation-harness)
# [Open LLM Leaderboard Evaluation Results](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard)
Detailed results can be found [here](https://huggingface.co/datasets/open-llm-leaderboard/details_janhq__Mistral-7B-Instruct-v0.2-DARE)

|             Metric              |Value|
|---------------------------------|----:|
|Avg.                             |55.84|
|AI2 Reasoning Challenge (25-Shot)|61.95|
|HellaSwag (10-Shot)              |75.62|
|MMLU (5-Shot)                    |49.99|
|TruthfulQA (0-shot)              |54.36|
|Winogrande (5-shot)              |74.98|
|GSM8k (5-shot)                   |18.12|