Text Classification
PEFT
Safetensors
English
npm
supply-chain-security
malware-detection
malicious-packages
security
triage
typed-decisions
lora
qlora
Instructions to use Takzen/jay-4b with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- PEFT
How to use Takzen/jay-4b with PEFT:
from peft import PeftModel from transformers import AutoModelForCausalLM base_model = AutoModelForCausalLM.from_pretrained("Qwen/Qwen3.5-4B") model = PeftModel.from_pretrained(base_model, "Takzen/jay-4b") - Notebooks
- Google Colab
- Kaggle
Download CALIBRATION.json from Takzen/jay-4b: direct link, hf CLI and curl.
- Browser
- Download file 420 Bytes
-
https://huggingface.co/Takzen/jay-4b/resolve/main/CALIBRATION.json
- Command line
-
hf download hf://Takzen/jay-4b/CALIBRATION.json
-
curl -L -o CALIBRATION.json https://huggingface.co/Takzen/jay-4b/resolve/main/CALIBRATION.json
420 Bytes
| { | |
| "temperature": { | |
| "choice": 1.231 | |
| }, | |
| "template": "decision-v1", | |
| "adapter": "artifacts/pilot2/adapter", | |
| "tta": false, | |
| "panel": { | |
| "n": 974, | |
| "skipped_long": 4, | |
| "families": { | |
| "verdict": 650, | |
| "origin": 328 | |
| } | |
| }, | |
| "by_type": { | |
| "choice": { | |
| "n": 974, | |
| "nll_before": 0.0714, | |
| "nll_after": 0.0679, | |
| "ece_before": 0.0129, | |
| "ece_after": 0.0082 | |
| } | |
| } | |
| } |