Zero-Shot Classification
Laya
Safetensors
GGUF
Hebrew
hebrew
decision-model
calibrated
scam-detection
routing
Instructions to use BrainboxAI/nitzotz with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Laya
How to use BrainboxAI/nitzotz with Laya:
# No code snippets available yet for this library. # To use this model, check the repository files and the library's documentation. # Want to help? PRs adding snippets are welcome at: # https://github.com/huggingface/huggingface.js
- Notebooks
- Google Colab
- Kaggle
Download calibration.json from BrainboxAI/nitzotz: direct link, hf CLI and curl.
- Browser
- Download file 1.47 kB
-
https://huggingface.co/BrainboxAI/nitzotz/resolve/main/calibration.json
- Command line
-
hf download hf://BrainboxAI/nitzotz/calibration.json
-
curl -L -o calibration.json https://huggingface.co/BrainboxAI/nitzotz/resolve/main/calibration.json
1.47 kB
| { | |
| "temperatures_per_question_type": { | |
| "choice": 1.0971, | |
| "score": 1.139, | |
| "noul": 1.162 | |
| }, | |
| "temperature_fit_items": { | |
| "choice": 2533, | |
| "score": 261, | |
| "noul": 2031 | |
| }, | |
| "note": "The temperatures are already applied by laya (rl_agent_config.json 'temperature') and baked into the GGUF files. They were fitted on held-out training items, never on the test sets.", | |
| "scam_threshold": { | |
| "question": "noul, 'is this message a scam'", | |
| "threshold": 0.4861, | |
| "chosen_on": "1188 held-out training messages, never the test", | |
| "held_out_acc_at_threshold": 0.979, | |
| "held_out_pools": { | |
| "held-out warnings about scams, short scams without a link and look-alike messages": { | |
| "n": 188, | |
| "acc_at_threshold": 0.984 | |
| }, | |
| "held-out hard scam and legitimate messages": { | |
| "n": 300, | |
| "acc_at_threshold": 0.9933 | |
| }, | |
| "held-out spam messages": { | |
| "n": 700, | |
| "acc_at_threshold": 0.9714 | |
| } | |
| }, | |
| "test_acc_at_0.5": 0.9195, | |
| "test_acc_at_threshold": 0.9228, | |
| "test_hard_acc_at_0.5": 0.8308, | |
| "test_hard_acc_at_threshold": 0.8308, | |
| "test_auc": 0.969 | |
| }, | |
| "ramzor_example_thresholds": { | |
| "green_below": 0.35, | |
| "red_from": 0.49, | |
| "zones_on_test": { | |
| "g": { | |
| "n": 207, | |
| "share": 0.6946308724832215, | |
| "scam": 13, | |
| "legit": 194 | |
| }, | |
| "y": { | |
| "n": 9, | |
| "share": 0.030201342281879196, | |
| "scam": 2, | |
| "legit": 7 | |
| }, | |
| "r": { | |
| "n": 82, | |
| "share": 0.2751677852348993, | |
| "scam": 73, | |
| "legit": 9 | |
| } | |
| } | |
| } | |
| } |