Datasets:
Add task_categories to metadata (#1)
Browse files- Add task_categories to metadata (c8f8803002972460bc70e53e121e48c201ee9b90)
Co-authored-by: Niels Rogge <nielsr@users.noreply.huggingface.co>
README.md
CHANGED
|
@@ -1,4 +1,7 @@
|
|
| 1 |
---
|
|
|
|
|
|
|
|
|
|
| 2 |
dataset_info:
|
| 3 |
- config_name: gemma-2-9b-it
|
| 4 |
features:
|
|
@@ -281,8 +284,8 @@ configs:
|
|
| 281 |
path: qwen2.5-7b-instruct/validation-*
|
| 282 |
- split: test
|
| 283 |
path: qwen2.5-7b-instruct/test-*
|
| 284 |
-
license: other
|
| 285 |
---
|
|
|
|
| 286 |
# BYOD: pretokenized 256-token training profiles
|
| 287 |
|
| 288 |
Paper: [BYOD: Build Your Own DLM — Efficient Autoregressive-to-Diffusion Language Model Conversion](https://arxiv.org/abs/2610.41574)
|
|
@@ -315,4 +318,4 @@ Comply with each source's terms and the applicable parent-model/tokenizer terms.
|
|
| 315 |
|
| 316 |
Sources include Clean-Instruct-3M (no license declared on its source card), Tulu 3
|
| 317 |
(ODC-By), Alpaca-GPT4/Alpaca (CC BY-NC 4.0), MMLU/Orca-Math/GSM8K/OpenCoder
|
| 318 |
-
(MIT), HellaSwag (MIT in its project), ARC (CC BY-SA 4.0), and MBPP (CC BY 4.0).
|
|
|
|
| 1 |
---
|
| 2 |
+
license: other
|
| 3 |
+
task_categories:
|
| 4 |
+
- text-generation
|
| 5 |
dataset_info:
|
| 6 |
- config_name: gemma-2-9b-it
|
| 7 |
features:
|
|
|
|
| 284 |
path: qwen2.5-7b-instruct/validation-*
|
| 285 |
- split: test
|
| 286 |
path: qwen2.5-7b-instruct/test-*
|
|
|
|
| 287 |
---
|
| 288 |
+
|
| 289 |
# BYOD: pretokenized 256-token training profiles
|
| 290 |
|
| 291 |
Paper: [BYOD: Build Your Own DLM — Efficient Autoregressive-to-Diffusion Language Model Conversion](https://arxiv.org/abs/2610.41574)
|
|
|
|
| 318 |
|
| 319 |
Sources include Clean-Instruct-3M (no license declared on its source card), Tulu 3
|
| 320 |
(ODC-By), Alpaca-GPT4/Alpaca (CC BY-NC 4.0), MMLU/Orca-Math/GSM8K/OpenCoder
|
| 321 |
+
(MIT), HellaSwag (MIT in its project), ARC (CC BY-SA 4.0), and MBPP (CC BY 4.0).
|