Instructions to use ed-00/KERMIT with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Adapters
How to use ed-00/KERMIT with Adapters:
from adapters import AutoAdapterModel model = AutoAdapterModel.from_pretrained("undefined") model.load_adapter("ed-00/KERMIT", set_active=True) - Notebooks
- Google Colab
- Kaggle
| license: llama2 | |
| datasets: | |
| - ed-00/SQUAD-Extractive-Instruction | |
| - Open-Orca/OpenOrca | |
| - Open-Orca/FLAN | |
| - ed-00/SQUAD_GPT4 | |
| language: | |
| - en | |
| library_name: adapter-transformers | |
| pipeline_tag: question-answering | |
| # KERMIT: Knowledge Extractive and Reasoning Model usIng Transformers | |
| ## Overview | |
| KERMIT (Knowledge Extractive and Reasoning Model usIng Transformers) is an advanced language model designed to perform complex reasoning tasks by leveraging the power of transformers. This model aims to enhance the ability to extract and reason over knowledge, making it particularly suitable for tasks such as question answering, summarization, and knowledge-based inference. | |
| (https://hig.diva-portal.org/smash/record.jsf?pid=diva2%3A1872915&dswid=-2622)[papper] |