Instructions to use sectionme/granite-clm-g2 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use sectionme/granite-clm-g2 with Transformers:
# pip install -U transformers accelerate # Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("sectionme/granite-clm-g2", device_map="auto") - Notebooks
- Google Colab
- Kaggle
Download granite_encoder.json from sectionme/granite-clm-g2: direct link, hf CLI and curl.
- Browser
- Download file 445 Bytes
-
https://huggingface.co/sectionme/granite-clm-g2/resolve/main/granite_encoder.json
- Command line
-
hf download hf://sectionme/granite-clm-g2/granite_encoder.json
-
curl -L -o granite_encoder.json https://huggingface.co/sectionme/granite-clm-g2/resolve/main/granite_encoder.json
445 Bytes
| { | |
| "encoder_id": "ibm-granite/granite-embedding-97m-multilingual-r2", | |
| "encoder_source": "ibm-granite/granite-embedding-97m-multilingual-r2+best.pt", | |
| "encoder_descriptor_digest": "be69086397edef79eecc0404478d0349e8a7485677b3ea78c56706d97a188757", | |
| "dimension": 384, | |
| "opset": 17, | |
| "pooling": "masked-mean-then-l2, applied by the runtime", | |
| "dtype": "f32", | |
| "note": "output is token-level; the runtime pools. Do not pool in the graph." | |
| } |