main-database/CodeUp.yaml at preview · Language-Technology-Assessment/main-database · GitHub

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
---
# Thank you for contributing!
# In filling out this yaml file, please follow the criteria as described here:
# https://osai-index.eu/contribute

# You're free to build on this work and reuse the data. It is licensed under CC-BY 4.0, with the
# stipulation that attribution should come in the form of a link to https://osai-index.eu/
# and a citation to the peer-reviewed paper in which the dataset & criteria were published:

# Liesenfeld, A. and Dingemanse, M., 2024. Rethinking open source generative AI: open-washing and the EU AI Act. In Proceedings of the 2024 ACM Conference on Fairness, Accountability, and Transparency (pp. 1774-1787).

# Organization tags:
# - National origin: China
# - Contributor type: Academic (University)

system:
    name: CodeUp
    link: https://huggingface.co/juyongjiang/CodeUp-Llama-3-8B
    type: code
    performanceclass: full
    basemodelname: Meta-Llama-3-8B
    endmodelname: CodeUp-Llama-3-8B
    endmodellicense: Stable-Diffusion-OpenRail++-M License and Apache-2.0 and CC-BY-NC-4.0
    releasedate: 2024-12
    notes: Multilingual code generation model.

org:
    name: John
    link: https://github.com/juyongjiang
    notes: PhD student at HKUST.

# availability:
datasources_basemodel:
    class: closed
    link:
    notes: Data nowhere disclosed or documented, and described only in the vaguest terms in a release blog post

datasources_endmodel:
    class: open
    link: ["https://huggingface.co/datasets/WizardLMTeam/WizardLM_evol_instruct_V2_196k", "https://huggingface.co/datasets/meta-math/MetaMathQA", "https://huggingface.co/datasets/theblackcat102/evol-codealpaca-v1"]
    notes: Datasets used for tuning published on HuggingFace.

weights_basemodel:
    class: partial
    link: https://ai.meta.com/resources/models-and-libraries/llama-downloads/
    notes: Download only after requesting access; requires signing a consent form

weights_endmodel:
    class: partial
    link: https://huggingface.co/juyongjiang/CodeUp-Llama-3-8B
    notes: Model available on HuggingFace.

trainingcode:
    class: partial
    link: ["https://github.com/meta-llama/llama3", "https://github.com/juyongjiang/CodeUp"]
    notes: Base model repository only offers minimal code. End model repo contains some code.

# documentation:
code:
    class: partial
    link: https://github.com/juyongjiang/CodeUp/blob/master/README.md
    notes: End model repo documented.

hardware_architecture:
    class: partial
    link: https://ai.meta.com/research/publications/llama-2-open-foundation-and-fine-tuned-chat-models/
    notes: Architecture sketched in glossy blog post.

preprint:
    class: partial
    link: https://arxiv.org/pdf/2406.00515
    notes: Preprint exists for a very tangentially related document.

paper:
    class: closed
    link:
    notes: No peer-reviewed paper found.

modelcard:
    class: partial
    link: https://huggingface.co/meta-llama/CodeLlama-70b-Instruct-hf
    notes: Model card provides some information about training and inference, however mostly contains usage instructions.

datasheet:
    class: partial
    link: ["https://huggingface.co/datasets/WizardLMTeam/WizardLM_evol_instruct_V2_196k", "https://huggingface.co/datasets/meta-math/MetaMathQA", "https://huggingface.co/datasets/theblackcat102/evol-codealpaca-v1"]
    notes: Datasheets found with various levels of detail.

# access:
package:
    class: closed
    link:
    notes: No package found.

api:
    class: closed
    link:
    notes: No API found.
    metaprompt: closed

licenses:
    class: partial
    link: https://huggingface.co/juyongjiang/CodeUp-Llama-3-8B
    notes: Stable-Diffusion-OpenRail++-M License and Apache-2.0 and CC-BY-NC-4.0. Not all OSI-approved licenses.