File size: 3,896 Bytes
11bd5eb
 
 
 
 
 
f822f93
11bd5eb
 
 
 
 
 
 
 
 
 
 
 
 
9c072ee
11bd5eb
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
c5df3ea
11bd5eb
 
 
bc72262
 
 
 
 
11bd5eb
 
837b122
 
 
 
11bd5eb
 
 
 
 
6676d6e
 
 
 
 
 
 
 
 
 
 
 
 
 
 
11bd5eb
5b46aff
 
 
 
 
 
f0ca9e5
 
5b46aff
 
6676d6e
 
5b46aff
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
---
dataset_info:
  features:
  - name: id
    dtype: string
  - name: author
    dtype: string
  - name: sha
    dtype: 'null'
  - name: created_at
    dtype: timestamp[us, tz=UTC]
  - name: last_modified
    dtype: 'null'
  - name: disabled
    dtype: 'null'
  - name: downloads
    dtype: int64
  - name: downloads_all_time
    dtype: 'null'
  - name: gated
    dtype: bool
  - name: gguf
    dtype: 'null'
  - name: inference
    dtype: 'null'
  - name: likes
    dtype: int64
  - name: library_name
    dtype: string
  - name: tags
    sequence: string
  - name: pipeline_tag
    dtype: string
  - name: mask_token
    dtype: 'null'
  - name: model_index
    dtype: 'null'
  - name: trending_score
    dtype: int64
  - name: architectures
    sequence: string
  - name: bos_token_id
    dtype: int64
  - name: eos_token_id
    dtype: int64
  - name: hidden_act
    dtype: string
  - name: hidden_size
    dtype: int64
  - name: initializer_range
    dtype: float64
  - name: intermediate_size
    dtype: int64
  - name: max_position_embeddings
    dtype: int64
  - name: model_type
    dtype: string
  - name: num_attention_heads
    dtype: int64
  - name: num_hidden_layers
    dtype: int64
  - name: num_key_value_heads
    dtype: int64
  - name: rms_norm_eps
    dtype: float64
  - name: rope_theta
    dtype: float64
  - name: sliding_window
    dtype: int64
  - name: tie_word_embeddings
    dtype: bool
  - name: torch_dtype
    dtype: string
  - name: transformers_version
    dtype: string
  - name: use_cache
    dtype: bool
  - name: vocab_size
    dtype: int64
  - name: attention_bias
    dtype: bool
  - name: attention_dropout
    dtype: float64
  - name: head_dim
    dtype: int64
  - name: mlp_bias
    dtype: bool
  - name: pretraining_tp
    dtype: int64
  - name: rope_scaling
    struct:
    - name: factor
      dtype: float64
    - name: original_max_position_embeddings
      dtype: float64
  splits:
  - name: raw
    num_bytes: 70119636
    num_examples: 129379
  download_size: 9132674
  dataset_size: 70119636
configs:
- config_name: default
  data_files:
  - split: raw
    path: data/raw-*
license: apache-2.0
task_categories:
- question-answering
language:
- en
- fr
tags:
- merge
- mergekit
- configs
- code
- automation
pretty_name: 'mergekit-configs: access all Hub architecture'
size_categories:
- 100K<n<1M
---

## Dataset Description
- **Repository:** https://huggingface.co/datasets/louisbrulenaudet/mergekit-configs
- **Leaderboard:** N/A
- **Point of Contact:** [Louis Brulé Naudet](mailto:[email protected])

<img src="assets/thumbnail.webp">

# MergeKit-configs: access all Hub architectures and automate your model merging process

This dataset facilitates the search for compatible architectures for model merging with MergeKit, streamlining the automation of high-performance merge searches. It provides a snapshot of the Hub’s configuration state, eliminating the need to manually open configuration files.

```python
import polars as pl

# Login using e.g. `huggingface-cli login` to access this dataset
df = pl.read_parquet('hf://datasets/louisbrulenaudet/mergekit-configs/data/raw-00000-of-00001.parquet')

result = (
  df.groupby(
    [
      "architectures",
      "hidden_size",
      "model_type",
      "intermediate_size"
    ]
  ).agg(
    pl.struct([pl.col("id")]).alias("models")
  )
)
```

## Citing & Authors

If you use this dataset in your research, please use the following BibTeX entry.

```BibTeX
@misc{HFforLegal2024,
  author =       {Louis Brulé Naudet},
  title =        {MergeKit-configs: access all Hub architectures and automate your model merging process},
  year =         {2024}
  howpublished = {\url{https://huggingface.co/datasets/louisbrulenaudet/mergekit-configs}},
}
```

## Feedback

If you have any feedback, please reach out at [[email protected]](mailto:[email protected]).