Upload README.md with huggingface_hub
Browse files
README.md
CHANGED
|
@@ -1,17 +1,12 @@
|
|
| 1 |
---
|
| 2 |
library_name: tokenizers
|
| 3 |
-
tags:
|
| 4 |
-
- kl3m
|
| 5 |
-
- kl3m-001
|
| 6 |
-
- alea
|
| 7 |
-
- legal
|
| 8 |
-
- financial
|
| 9 |
date: 2023-12-28
|
| 10 |
---
|
| 11 |
|
| 12 |
# kl3m-001-32k tokenizer
|
| 13 |
|
| 14 |
-
The `kl3m-001-32k` tokenizer is a domain-specific tokenizer trained on ~500B of financial and legal text from primarily-English sources.
|
| 15 |
|
| 16 |
This tokenizer was used for the first generation of KL3M embedding and generative models, including
|
| 17 |
`kl3m-170M`, `kl3m-1.7B`, `kl3m-embedding-001`, and `kl3m-embedding-002`.
|
|
@@ -33,7 +28,7 @@ Please see `kl3m-003-64k` for the next iteration of our research on domain-speci
|
|
| 33 |
|
| 34 |
### Model Description
|
| 35 |
|
| 36 |
-
The `kl3m-001-32k` tokenizer is a domain-specific tokenizer trained on ~500B of financial and legal text from primarily-English sources.
|
| 37 |
|
| 38 |
This tokenizer is notable for a number of reasons:
|
| 39 |
|
|
|
|
| 1 |
---
|
| 2 |
library_name: tokenizers
|
| 3 |
+
tags: ['kl3m', 'kl3m-001', 'alea', 'legal', 'financial']
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 4 |
date: 2023-12-28
|
| 5 |
---
|
| 6 |
|
| 7 |
# kl3m-001-32k tokenizer
|
| 8 |
|
| 9 |
+
The `kl3m-001-32k` tokenizer is a domain-specific tokenizer trained on ~500B tokens of financial and legal text from primarily-English sources.
|
| 10 |
|
| 11 |
This tokenizer was used for the first generation of KL3M embedding and generative models, including
|
| 12 |
`kl3m-170M`, `kl3m-1.7B`, `kl3m-embedding-001`, and `kl3m-embedding-002`.
|
|
|
|
| 28 |
|
| 29 |
### Model Description
|
| 30 |
|
| 31 |
+
The `kl3m-001-32k` tokenizer is a domain-specific tokenizer trained on ~500B tokens of financial and legal text from primarily-English sources.
|
| 32 |
|
| 33 |
This tokenizer is notable for a number of reasons:
|
| 34 |
|