Instructions to use hurtmongoose/tinyllama-lora-malicious-classifier with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- PEFT
How to use hurtmongoose/tinyllama-lora-malicious-classifier with PEFT:
from peft import PeftModel from transformers import AutoModelForSequenceClassification base_model = AutoModelForSequenceClassification.from_pretrained("TinyLlama/TinyLlama_v1.1") model = PeftModel.from_pretrained(base_model, "hurtmongoose/tinyllama-lora-malicious-classifier") - Transformers
How to use hurtmongoose/tinyllama-lora-malicious-classifier with Transformers:
# pip install -U transformers accelerate # Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("hurtmongoose/tinyllama-lora-malicious-classifier", device_map="auto") - Notebooks
- Google Colab
- Kaggle
Download eval_metrics.json from hurtmongoose/tinyllama-lora-malicious-classifier: direct link, hf CLI and curl.
- Browser
- Download file 1.18 kB
-
https://huggingface.co/hurtmongoose/tinyllama-lora-malicious-classifier/resolve/main/eval_metrics.json
- Command line
-
hf download hf://hurtmongoose/tinyllama-lora-malicious-classifier/eval_metrics.json
-
curl -L -o eval_metrics.json https://huggingface.co/hurtmongoose/tinyllama-lora-malicious-classifier/resolve/main/eval_metrics.json
1.18 kB
| { | |
| "eval_loss": 0.48330193758010864, | |
| "eval_accuracy": 0.8289269051321928, | |
| "eval_precision_weighted": 0.821952872098563, | |
| "eval_recall_weighted": 0.8289269051321928, | |
| "eval_f1_weighted": 0.8238609492009197, | |
| "eval_MCC": 0.7203450380787764, | |
| "eval_balanced_accuracy": 0.7723728109308511, | |
| "eval_macro_FNR": 0.227627189069149, | |
| "eval_macro_FPR": 0.08973774534029767, | |
| "eval_macro_Specificity": 0.9102622546597022, | |
| "eval_per_class": { | |
| "jailbreaking": { | |
| "TP": 228, | |
| "FP": 103, | |
| "FN": 161, | |
| "TN": 1437, | |
| "FNR": 0.4138817480719794, | |
| "FPR": 0.06688311688311688, | |
| "Specificity": 0.9331168831168831 | |
| }, | |
| "prompt injection": { | |
| "TP": 439, | |
| "FP": 112, | |
| "FN": 131, | |
| "TN": 1247, | |
| "FNR": 0.22982456140350876, | |
| "FPR": 0.08241353936718175, | |
| "Specificity": 0.9175864606328182 | |
| }, | |
| "unharmful": { | |
| "TP": 932, | |
| "FP": 115, | |
| "FN": 38, | |
| "TN": 844, | |
| "FNR": 0.03917525773195876, | |
| "FPR": 0.11991657977059438, | |
| "Specificity": 0.8800834202294057 | |
| } | |
| }, | |
| "eval_runtime": 44.2738, | |
| "eval_samples_per_second": 43.57, | |
| "eval_steps_per_second": 5.466, | |
| "epoch": 15.0 | |
| } |