File size: 7,270 Bytes
df53d7b
e5e8c1d
df53d7b
 
 
 
e5e8c1d
 
 
df53d7b
e5e8c1d
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
df53d7b
 
 
 
 
 
 
 
 
 
 
 
 
ea44990
 
df53d7b
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
e5e8c1d
 
 
 
 
 
 
 
 
 
 
 
 
 
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
---
license: cc-by-nc-4.0
library_name: transformers
tags:
- mergekit
- merge
base_model:
- WhiteRabbitNeo/Llama-3-WhiteRabbitNeo-8B-v2.0
- mlabonne/NeuralDaredevil-8B-abliterated
pipeline_tag: text-generation
model-index:
- name: llama-3-Nephilim-v1-8B
  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: 42.77
      name: strict accuracy
    source:
      url: https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=grimjim/llama-3-Nephilim-v1-8B
      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: 29.91
      name: normalized accuracy
    source:
      url: https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=grimjim/llama-3-Nephilim-v1-8B
      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: 8.16
      name: exact match
    source:
      url: https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=grimjim/llama-3-Nephilim-v1-8B
      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: 6.94
      name: acc_norm
    source:
      url: https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=grimjim/llama-3-Nephilim-v1-8B
      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: 10.64
      name: acc_norm
    source:
      url: https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=grimjim/llama-3-Nephilim-v1-8B
      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: 31.06
      name: accuracy
    source:
      url: https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=grimjim/llama-3-Nephilim-v1-8B
      name: Open LLM Leaderboard
---
# llama-3-Nephilim-v1-8B

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

Here we experiment with SLERP merger with the second model at very low weight (0.001) to modulate the output of the base model.

The base model was assembled to achieve high MMLU while avoiding refusals, while the additional model was trained specifically (apparently as a copilot) for offensive and defensive cybersecurity. Though neither model targeted roleplay as a use case, the resulting intelligence, acuity, and text generation of the merge is of interest. The merge is aggressively creative, within bounds.

Tested with temperature=1.0-1.2 and minP=0.01 along with a custom Instruct prompt geared toward reducing refusals during roleplay text generation without compromising overall model safety: [Llama 3 Instruct Direct](https://huggingface.co/debased-ai/SillyTavern-settings/tree/main/advanced_formatting/instruct_mode).

Care should be taken when using this model, as it is possible that harmful outputs could be generated. Given that this model is derivative, responsible use is further mandated by the WhiteRabbitNeo Usage Restrictions Extension to the Llama-3 License. This model is further subject to CC-BY-NC-4.0 by default, meaning that commercial use is restricted, barring an alternative licensing agreement.

- [GGUF quants (w/ llama-bpe pre-tokenizer)](https://huggingface.co/grimjim/llama-3-Nephilim-v1-8B-GGUF)

Built with Meta Llama 3.

# WhiteRabbitNeo Extension to Llama-3 Licence: Usage Restrictions

```
You agree not to use the Model or Derivatives of the Model:

-	In any way that violates any applicable national or international law or regulation or infringes upon the lawful rights and interests of any third party; 
-	For military use in any way;
-	For the purpose of exploiting, harming or attempting to exploit or harm minors in any way; 
-	To generate or disseminate verifiably false information and/or content with the purpose of harming others; 
-	To generate or disseminate inappropriate content subject to applicable regulatory requirements;
-	To generate or disseminate personal identifiable information without due authorization or for unreasonable use; 
-	To defame, disparage or otherwise harass others; 
-	For fully automated decision making that adversely impacts an individual's legal rights or otherwise creates or modifies a binding, enforceable obligation; 
-	For any use intended to or which has the effect of discriminating against or harming individuals or groups based on online or offline social behavior or known or predicted personal or personality characteristics; 
-	To exploit any of the vulnerabilities of a specific group of persons based on their age, social, physical or mental characteristics, in order to materially distort the behavior of a person pertaining to that group in a manner that causes or is likely to cause that person or another person physical or psychological harm; 
-	For any use intended to or which has the effect of discriminating against individuals or groups based on legally protected characteristics or categories.
```

## Merge Details
### Merge Method

This model was merged using the SLERP merge method.

### Models Merged

The following models were included in the merge:
* [WhiteRabbitNeo/Llama-3-WhiteRabbitNeo-8B-v2.0](https://huggingface.co/WhiteRabbitNeo/Llama-3-WhiteRabbitNeo-8B-v2.0)
* [mlabonne/NeuralDaredevil-8B-abliterated](https://huggingface.co/mlabonne/NeuralDaredevil-8B-abliterated)

### Configuration

The following YAML configuration was used to produce this model:

```yaml
slices:
  - sources:
    - model: mlabonne/NeuralDaredevil-8B-abliterated
      layer_range: [0,32]
    - model: WhiteRabbitNeo/Llama-3-WhiteRabbitNeo-8B-v2.0
      layer_range: [0,32]
merge_method: slerp
base_model: mlabonne/NeuralDaredevil-8B-abliterated
parameters:
  t:
    - value: 0.001
dtype: bfloat16

```

# [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_grimjim__llama-3-Nephilim-v1-8B)

|      Metric       |Value|
|-------------------|----:|
|Avg.               |21.58|
|IFEval (0-Shot)    |42.77|
|BBH (3-Shot)       |29.91|
|MATH Lvl 5 (4-Shot)| 8.16|
|GPQA (0-shot)      | 6.94|
|MuSR (0-shot)      |10.64|
|MMLU-PRO (5-shot)  |31.06|