goldfish-models
commited on
Commit
•
eb7784f
1
Parent(s):
8da917d
Upload README.md with huggingface_hub
Browse files
README.md
CHANGED
@@ -11,7 +11,7 @@ library_name: transformers
|
|
11 |
pipeline_tag: text-generation
|
12 |
tags:
|
13 |
- goldfish
|
14 |
-
|
15 |
---
|
16 |
|
17 |
# bcl_latn_full
|
@@ -22,7 +22,7 @@ The Goldfish models are trained primarily for comparability across languages and
|
|
22 |
|
23 |
Note: bcl_latn is an [individual language](https://iso639-3.sil.org/code_tables/639/data) code. Macrolanguage code bik_latn (Bikol) is included in Goldfish. Consider using that model depending on your use case.
|
24 |
|
25 |
-
All training and hyperparameter details are in our paper, [Goldfish: Monolingual Language Models for 350 Languages (Chang et al., 2024)](https://
|
26 |
|
27 |
Training code and sample usage: https://github.com/tylerachang/goldfish
|
28 |
|
@@ -32,6 +32,7 @@ Sample usage also in this Google Colab: [link](https://colab.research.google.com
|
|
32 |
|
33 |
To access all Goldfish model details programmatically, see https://github.com/tylerachang/goldfish/blob/main/model_details.json.
|
34 |
All models are trained with a [CLS] (same as [BOS]) token prepended, and a [SEP] (same as [EOS]) token separating sequences.
|
|
|
35 |
Details for this model specifically:
|
36 |
|
37 |
* Architecture: gpt2
|
@@ -58,5 +59,6 @@ If you use this model, please cite:
|
|
58 |
author={Chang, Tyler A. and Arnett, Catherine and Tu, Zhuowen and Bergen, Benjamin K.},
|
59 |
journal={Preprint},
|
60 |
year={2024},
|
|
|
61 |
}
|
62 |
```
|
|
|
11 |
pipeline_tag: text-generation
|
12 |
tags:
|
13 |
- goldfish
|
14 |
+
- arxiv:2408.10441
|
15 |
---
|
16 |
|
17 |
# bcl_latn_full
|
|
|
22 |
|
23 |
Note: bcl_latn is an [individual language](https://iso639-3.sil.org/code_tables/639/data) code. Macrolanguage code bik_latn (Bikol) is included in Goldfish. Consider using that model depending on your use case.
|
24 |
|
25 |
+
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).
|
26 |
|
27 |
Training code and sample usage: https://github.com/tylerachang/goldfish
|
28 |
|
|
|
32 |
|
33 |
To access all Goldfish model details programmatically, see https://github.com/tylerachang/goldfish/blob/main/model_details.json.
|
34 |
All models are trained with a [CLS] (same as [BOS]) token prepended, and a [SEP] (same as [EOS]) token separating sequences.
|
35 |
+
For best results, make sure that [CLS] is prepended to your input sequence (see sample usage linked above)!
|
36 |
Details for this model specifically:
|
37 |
|
38 |
* Architecture: gpt2
|
|
|
59 |
author={Chang, Tyler A. and Arnett, Catherine and Tu, Zhuowen and Bergen, Benjamin K.},
|
60 |
journal={Preprint},
|
61 |
year={2024},
|
62 |
+
url={https://www.arxiv.org/abs/2408.10441},
|
63 |
}
|
64 |
```
|