goldfish-models
commited on
Commit
•
c992136
1
Parent(s):
cf34874
Upload README.md with huggingface_hub
Browse files
README.md
CHANGED
@@ -3,8 +3,8 @@
|
|
3 |
license: apache-2.0
|
4 |
language:
|
5 |
- zho
|
6 |
-
- chi
|
7 |
- yue
|
|
|
8 |
datasets:
|
9 |
- cis-lmu/Glot500
|
10 |
- legacy-datasets/wikipedia
|
@@ -12,7 +12,7 @@ library_name: transformers
|
|
12 |
pipeline_tag: text-generation
|
13 |
tags:
|
14 |
- goldfish
|
15 |
-
|
16 |
---
|
17 |
|
18 |
# yue_hant_full
|
@@ -23,7 +23,7 @@ The Goldfish models are trained primarily for comparability across languages and
|
|
23 |
|
24 |
Note: yue_hant is an [individual language](https://iso639-3.sil.org/code_tables/639/data) code. Macrolanguage code zho_hant (Chinese) is included in Goldfish. Consider using that model depending on your use case.
|
25 |
|
26 |
-
All training and hyperparameter details are in our paper, [Goldfish: Monolingual Language Models for 350 Languages (Chang et al., 2024)](https://
|
27 |
|
28 |
Training code and sample usage: https://github.com/tylerachang/goldfish
|
29 |
|
@@ -33,6 +33,7 @@ Sample usage also in this Google Colab: [link](https://colab.research.google.com
|
|
33 |
|
34 |
To access all Goldfish model details programmatically, see https://github.com/tylerachang/goldfish/blob/main/model_details.json.
|
35 |
All models are trained with a [CLS] (same as [BOS]) token prepended, and a [SEP] (same as [EOS]) token separating sequences.
|
|
|
36 |
Details for this model specifically:
|
37 |
|
38 |
* Architecture: gpt2
|
@@ -60,5 +61,6 @@ If you use this model, please cite:
|
|
60 |
author={Chang, Tyler A. and Arnett, Catherine and Tu, Zhuowen and Bergen, Benjamin K.},
|
61 |
journal={Preprint},
|
62 |
year={2024},
|
|
|
63 |
}
|
64 |
```
|
|
|
3 |
license: apache-2.0
|
4 |
language:
|
5 |
- zho
|
|
|
6 |
- yue
|
7 |
+
- chi
|
8 |
datasets:
|
9 |
- cis-lmu/Glot500
|
10 |
- legacy-datasets/wikipedia
|
|
|
12 |
pipeline_tag: text-generation
|
13 |
tags:
|
14 |
- goldfish
|
15 |
+
- arxiv:2408.10441
|
16 |
---
|
17 |
|
18 |
# yue_hant_full
|
|
|
23 |
|
24 |
Note: yue_hant is an [individual language](https://iso639-3.sil.org/code_tables/639/data) code. Macrolanguage code zho_hant (Chinese) is included in Goldfish. Consider using that model depending on your use case.
|
25 |
|
26 |
+
All training and hyperparameter details are in our paper, [Goldfish: Monolingual Language Models for 350 Languages (Chang et al., 2024)](https://www.arxiv.org/abs/2408.10441).
|
27 |
|
28 |
Training code and sample usage: https://github.com/tylerachang/goldfish
|
29 |
|
|
|
33 |
|
34 |
To access all Goldfish model details programmatically, see https://github.com/tylerachang/goldfish/blob/main/model_details.json.
|
35 |
All models are trained with a [CLS] (same as [BOS]) token prepended, and a [SEP] (same as [EOS]) token separating sequences.
|
36 |
+
For best results, make sure that [CLS] is prepended to your input sequence (see sample usage linked above)!
|
37 |
Details for this model specifically:
|
38 |
|
39 |
* Architecture: gpt2
|
|
|
61 |
author={Chang, Tyler A. and Arnett, Catherine and Tu, Zhuowen and Bergen, Benjamin K.},
|
62 |
journal={Preprint},
|
63 |
year={2024},
|
64 |
+
url={https://www.arxiv.org/abs/2408.10441},
|
65 |
}
|
66 |
```
|