goldfish-models
commited on
Commit
•
27711ca
1
Parent(s):
2274a55
Upload README.md with huggingface_hub
Browse files
README.md
CHANGED
@@ -9,7 +9,7 @@ library_name: transformers
|
|
9 |
pipeline_tag: text-generation
|
10 |
tags:
|
11 |
- goldfish
|
12 |
-
|
13 |
---
|
14 |
|
15 |
# bul_cyrl_100mb
|
@@ -20,7 +20,7 @@ The Goldfish models are trained primarily for comparability across languages and
|
|
20 |
|
21 |
Note: bul_cyrl is an [individual language](https://iso639-3.sil.org/code_tables/639/data) code. It is not contained in any macrolanguage codes contained in Goldfish (for script cyrl).
|
22 |
|
23 |
-
All training and hyperparameter details are in our paper, [Goldfish: Monolingual Language Models for 350 Languages (Chang et al., 2024)](https://
|
24 |
|
25 |
Training code and sample usage: https://github.com/tylerachang/goldfish
|
26 |
|
@@ -30,6 +30,7 @@ Sample usage also in this Google Colab: [link](https://colab.research.google.com
|
|
30 |
|
31 |
To access all Goldfish model details programmatically, see https://github.com/tylerachang/goldfish/blob/main/model_details.json.
|
32 |
All models are trained with a [CLS] (same as [BOS]) token prepended, and a [SEP] (same as [EOS]) token separating sequences.
|
|
|
33 |
Details for this model specifically:
|
34 |
|
35 |
* Architecture: gpt2
|
@@ -55,5 +56,6 @@ If you use this model, please cite:
|
|
55 |
author={Chang, Tyler A. and Arnett, Catherine and Tu, Zhuowen and Bergen, Benjamin K.},
|
56 |
journal={Preprint},
|
57 |
year={2024},
|
|
|
58 |
}
|
59 |
```
|
|
|
9 |
pipeline_tag: text-generation
|
10 |
tags:
|
11 |
- goldfish
|
12 |
+
- arxiv:2408.10441
|
13 |
---
|
14 |
|
15 |
# bul_cyrl_100mb
|
|
|
20 |
|
21 |
Note: bul_cyrl is an [individual language](https://iso639-3.sil.org/code_tables/639/data) code. It is not contained in any macrolanguage codes contained in Goldfish (for script cyrl).
|
22 |
|
23 |
+
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).
|
24 |
|
25 |
Training code and sample usage: https://github.com/tylerachang/goldfish
|
26 |
|
|
|
30 |
|
31 |
To access all Goldfish model details programmatically, see https://github.com/tylerachang/goldfish/blob/main/model_details.json.
|
32 |
All models are trained with a [CLS] (same as [BOS]) token prepended, and a [SEP] (same as [EOS]) token separating sequences.
|
33 |
+
For best results, make sure that [CLS] is prepended to your input sequence (see sample usage linked above)!
|
34 |
Details for this model specifically:
|
35 |
|
36 |
* Architecture: gpt2
|
|
|
56 |
author={Chang, Tyler A. and Arnett, Catherine and Tu, Zhuowen and Bergen, Benjamin K.},
|
57 |
journal={Preprint},
|
58 |
year={2024},
|
59 |
+
url={https://www.arxiv.org/abs/2408.10441},
|
60 |
}
|
61 |
```
|