Text Classification
Transformers
Safetensors
English
qwen3_5_text
text-generation
system-one
typed-decisions
decision-model
calibrated-probabilities
knowledge-distillation
jev
noul
choice
score
lora
qwen3_5
dual-head
vllm
Eval Results (legacy)
Instructions to use autotrust/JEV with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use autotrust/JEV with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-classification", model="autotrust/JEV")# Load model directly from transformers import AutoTokenizer, AutoModelForCausalLM tokenizer = AutoTokenizer.from_pretrained("autotrust/JEV") model = AutoModelForCausalLM.from_pretrained("autotrust/JEV", device_map="auto") - Notebooks
- Google Colab
- Kaggle
| #!/usr/bin/env python3 | |
| """Upload an export bundle (+ model card) to the Hugging Face Hub. | |
| python3 scripts/push_hf.py --export exports/jev-judge-qwen35-9b --repo autotrust/JEV --dry-run | |
| python3 scripts/push_hf.py --export exports/jev-judge-qwen35-9b --repo autotrust/JEV --private | |
| """ | |
| from __future__ import annotations | |
| import argparse | |
| import os | |
| import yaml | |
| def validate_card(path: str) -> dict: | |
| text = open(path, encoding="utf-8").read() | |
| if not text.startswith("---\n"): | |
| raise SystemExit("README.md has no YAML front matter") | |
| fm = text.split("---\n", 2)[1] | |
| meta = yaml.safe_load(fm) | |
| for k in ("license", "base_model", "library_name", "pipeline_tag", "tags"): | |
| if k not in meta: | |
| raise SystemExit(f"model card front matter missing `{k}`") | |
| return meta | |
| def main() -> None: | |
| ap = argparse.ArgumentParser() | |
| ap.add_argument("--export", required=True) | |
| ap.add_argument("--repo", required=True, help="e.g. autotrust/JEV") | |
| ap.add_argument("--private", action="store_true") | |
| ap.add_argument("--dry-run", action="store_true") | |
| ap.add_argument("--message", default="Upload JEV v0.7.0 (Qwen3.5-9B distilled from Jev 1.13)") | |
| args = ap.parse_args() | |
| card = os.path.join(args.export, "README.md") | |
| meta = validate_card(card) | |
| files = sorted(os.listdir(args.export)) | |
| total = sum(os.path.getsize(os.path.join(args.export, f)) for f in files) | |
| print(f"repo: {args.repo} ({'private' if args.private else 'public'})") | |
| print(f"card front matter: license={meta['license']} base_model={meta['base_model']} pipeline_tag={meta['pipeline_tag']} tags={len(meta['tags'])}") | |
| print(f"files ({total/1e9:.2f} GB):") | |
| for f in files: | |
| print(f" {os.path.getsize(os.path.join(args.export, f))/1e9:8.3f} GB {f}") | |
| if args.dry_run: | |
| print("dry run — nothing uploaded") | |
| return | |
| from huggingface_hub import HfApi | |
| api = HfApi() | |
| who = api.whoami() | |
| print("authenticated as", who.get("name"), "orgs:", [o.get("name") for o in who.get("orgs", [])]) | |
| api.create_repo(args.repo, repo_type="model", private=args.private, exist_ok=True) | |
| url = api.upload_folder(repo_id=args.repo, repo_type="model", folder_path=args.export, commit_message=args.message, | |
| ignore_patterns=["*.log", "__pycache__"], | |
| delete_patterns=["model-*.safetensors", "model.safetensors.index.json", "code/**"]) | |
| print("uploaded ->", url) | |
| print(f"https://huggingface.co/{args.repo}") | |
| if __name__ == "__main__": | |
| main() | |