Text Classification
Transformers
Safetensors
modernbert
cross-encoder
context-compression
supercompress
neural-keep
text-embeddings-inference
Instructions to use arjunkshah21/sc-keep-crossencoder-v4-large with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use arjunkshah21/sc-keep-crossencoder-v4-large with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-classification", model="arjunkshah21/sc-keep-crossencoder-v4-large")# pip install -U transformers accelerate # Load model directly from transformers import AutoTokenizer, AutoModelForSequenceClassification tokenizer = AutoTokenizer.from_pretrained("arjunkshah21/sc-keep-crossencoder-v4-large") model = AutoModelForSequenceClassification.from_pretrained("arjunkshah21/sc-keep-crossencoder-v4-large", device_map="auto") - Notebooks
- Google Colab
- Kaggle
Download inference.json from arjunkshah21/sc-keep-crossencoder-v4-large: direct link, hf CLI and curl.
- Browser
- Download file 327 Bytes
-
https://huggingface.co/arjunkshah21/sc-keep-crossencoder-v4-large/resolve/main/inference.json
- Command line
-
hf download hf://arjunkshah21/sc-keep-crossencoder-v4-large/inference.json
-
curl -L -o inference.json https://huggingface.co/arjunkshah21/sc-keep-crossencoder-v4-large/resolve/main/inference.json
327 Bytes
| { | |
| "model_dir": "checkpoints/sc-keep-crossencoder-v4-large", | |
| "task": "query_conditioned_line_keep", | |
| "keep_threshold": 0.95, | |
| "keep_floor": 0.98, | |
| "max_len": 128, | |
| "base_model": "answerdotai/ModernBERT-large", | |
| "approx_params_M": 395, | |
| "notes": "v4 large MPS \u2014 P(keep)>=keep_threshold; gated on launch_bench_30" | |
| } |