File size: 6,582 Bytes
1273b35
9d51581
1273b35
 
 
 
979334b
9d51581
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
979334b
 
9d51581
 
 
 
 
 
 
 
 
 
 
 
 
 
979334b
 
9d51581
 
 
 
 
 
 
 
 
 
 
 
 
 
979334b
 
9d51581
 
 
 
 
 
 
 
 
 
 
 
 
 
979334b
 
9d51581
 
 
 
 
 
 
 
 
 
 
 
 
 
979334b
 
9d51581
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
979334b
 
9d51581
1273b35
c78b4be
7147e4b
f4d7234
7147e4b
c78b4be
 
 
 
 
 
 
 
 
8701785
c78b4be
8701785
c78b4be
 
 
 
 
 
 
7147e4b
1273b35
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
9d51581
 
 
 
 
 
 
 
 
 
 
 
979334b
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
---
license: apache-2.0
library_name: transformers
tags:
- mergekit
- merge
- story-writing
base_model:
- nbeerbower/mistral-nemo-gutenberg-12B-v4
- Gryphe/Pantheon-RP-1.6.1-12b-Nemo
- anthracite-org/magnum-v4-12b
model-index:
- name: Pans_Gutenbergum_V0.1
  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: 30.97
      name: strict accuracy
    source:
      url: >-
        https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=Triangle104/Pans_Gutenbergum_V0.1
      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: 36.08
      name: normalized accuracy
    source:
      url: >-
        https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=Triangle104/Pans_Gutenbergum_V0.1
      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: 10.27
      name: exact match
    source:
      url: >-
        https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=Triangle104/Pans_Gutenbergum_V0.1
      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: 9.73
      name: acc_norm
    source:
      url: >-
        https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=Triangle104/Pans_Gutenbergum_V0.1
      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: 16.33
      name: acc_norm
    source:
      url: >-
        https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=Triangle104/Pans_Gutenbergum_V0.1
      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: 29.96
      name: accuracy
    source:
      url: >-
        https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=Triangle104/Pans_Gutenbergum_V0.1
      name: Open LLM Leaderboard
---

---
This seems to be working a bit better than my last merge attempts:
-

USER: Write a story about a Hobbit called Bilbo Baggins.

ASSISTANT:

Bildo Baggins lived in the quiet Shire village of Hobbiton. He was known far and wide for his love of the simple things in life - good food, fine ale, and quality time spent with friends.

One day, as he sat by the fireside in the Green Dragon Inn, Bildo overheard a group of hobbits discussing their latest hobby: collecting unique and interesting names. Intrigued, Bildo decided to embark on this new pastime himself.

Over the next few weeks, Bildo made it his mission to meet as many interesting characters as possible. He traveled far and wide across Middle-earth, chatting with elves, dwarves, men, and even a few peculiar creatures along the way. Each encounter yielded a new name for his growing collection.

As word of Bildo's quest spread, hobbits from all over began to seek him out, eager to share their own unusual names or to hear about the ones he had gathered. Before long, Bildo had amassed an impressive list - from the whimsical (Pippa Pincherly) to the downright peculiar (Thorn Toecracker).

But Bildo's true passion was not just collecting names, but understanding the stories behind them. He would spend hours listening intently as his new acquaintances recounted tales of adventure and misadventure, weaving a tapestry of Middle-earth's history through the lens of its inhabitants' names.

As time went on, Bildo became something of a local celebrity in Hobbiton. Hobbits would gather at his doorstep to hear him recount stories and share the names he had collected. And so, Bildo Baggins found his true calling - not as an adventurer or scholar, but as a keeper of Middle-earth's oral history, one name at a time.

And that, my dear reader, is the tale of how Bildo Baggins became known far and wide for his love of unusual names and the stories behind them. May your own journeys be filled with wonder and curiosity, just like those of our friend Bildo!

---
# 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 [TIES](https://arxiv.org/abs/2306.01708) merge method using [anthracite-org/magnum-v4-12b](https://huggingface.co/anthracite-org/magnum-v4-12b) as a base.

### Models Merged

The following models were included in the merge:
* [nbeerbower/mistral-nemo-gutenberg-12B-v4](https://huggingface.co/nbeerbower/mistral-nemo-gutenberg-12B-v4)
* [Gryphe/Pantheon-RP-1.6.1-12b-Nemo](https://huggingface.co/Gryphe/Pantheon-RP-1.6.1-12b-Nemo)

### Configuration

The following YAML configuration was used to produce this model:

```yaml
models:
  - model: anthracite-org/magnum-v4-12b
    #no parameters necessary for base model
  - model: Gryphe/Pantheon-RP-1.6.1-12b-Nemo
    parameters:
      density: 0.5
      weight: 0.5
  - model: nbeerbower/mistral-nemo-gutenberg-12B-v4
    parameters:
      density: 0.5
      weight: 0.5

merge_method: ties
base_model: anthracite-org/magnum-v4-12b
parameters:
  normalize: false
  int8_mask: true
dtype: float16
```
# [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/Triangle104__Pans_Gutenbergum_V0.1-details)

|      Metric       |Value|
|-------------------|----:|
|Avg.               |22.23|
|IFEval (0-Shot)    |30.97|
|BBH (3-Shot)       |36.08|
|MATH Lvl 5 (4-Shot)|10.27|
|GPQA (0-shot)      | 9.73|
|MuSR (0-shot)      |16.33|
|MMLU-PRO (5-shot)  |29.96|