Update README.md
Browse files
README.md
CHANGED
|
@@ -27,6 +27,8 @@ Currently, we open-source checkpoints of Cuckoos that are pre-trained on:
|
|
| 27 |
|
| 28 |
4) Cuckoo-C4-Rainbow + Multiple NER Datasets, WizardLM Dataset, Multiple Choice QA Datasets, MMLU, SQuAD, DROP, MNLI, SNLI. ([Cuckoo-C4-Super-Rainbow](https://huggingface.co/KomeijiForce/Cuckoo-C4-Super-Rainbow) 🦸🌈🐦🛠️)
|
| 29 |
|
|
|
|
|
|
|
| 30 |
## Performance Demonstration 🚀
|
| 31 |
|
| 32 |
Begin your journey with Cuckoo to experience unimaginable adaptation efficiency for all kinds of IE tasks!
|
|
|
|
| 27 |
|
| 28 |
4) Cuckoo-C4-Rainbow + Multiple NER Datasets, WizardLM Dataset, Multiple Choice QA Datasets, MMLU, SQuAD, DROP, MNLI, SNLI. ([Cuckoo-C4-Super-Rainbow](https://huggingface.co/KomeijiForce/Cuckoo-C4-Super-Rainbow) 🦸🌈🐦🛠️)
|
| 29 |
|
| 30 |
+
* This dataset only contains the first 1M split for C4 NTE, you can convert the full C4 (and other corpus) with [this script](https://github.com/KomeijiForce/Cuckoo/blob/main/nte_data_collection.py).
|
| 31 |
+
|
| 32 |
## Performance Demonstration 🚀
|
| 33 |
|
| 34 |
Begin your journey with Cuckoo to experience unimaginable adaptation efficiency for all kinds of IE tasks!
|