Instructions to use itsalissonsilva/SecPaper-Selector-4B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- PEFT
How to use itsalissonsilva/SecPaper-Selector-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, "itsalissonsilva/SecPaper-Selector-4B") - Notebooks
- Google Colab
- Kaggle
Download cpu_baseline_evaluation.json from itsalissonsilva/SecPaper-Selector-4B: direct link, hf CLI and curl.
- Browser
- Download file 1.43 kB
-
https://huggingface.co/itsalissonsilva/SecPaper-Selector-4B/resolve/main/cpu_baseline_evaluation.json
- Command line
-
hf download hf://itsalissonsilva/SecPaper-Selector-4B/cpu_baseline_evaluation.json
-
curl -L -o cpu_baseline_evaluation.json https://huggingface.co/itsalissonsilva/SecPaper-Selector-4B/resolve/main/cpu_baseline_evaluation.json
1.43 kB
| { | |
| "protocol": { | |
| "families": [ | |
| "tfidf_pair_logistic", | |
| "lexical_histgradient" | |
| ], | |
| "selection": "maximum validation balanced exact-set accuracy, then accuracy", | |
| "thresholds": [ | |
| 0.1, | |
| 0.2, | |
| 0.3, | |
| 0.4, | |
| 0.5, | |
| 0.6, | |
| 0.7, | |
| 0.8, | |
| 0.9 | |
| ], | |
| "caps": [ | |
| 1, | |
| 2, | |
| 3, | |
| 6 | |
| ], | |
| "test_used_for_selection": false, | |
| "added_after_native_result": true, | |
| "seed": 2026110607 | |
| }, | |
| "selected": { | |
| "family": 0, | |
| "threshold": 0.7, | |
| "cap": 1, | |
| "validation": { | |
| "n": 101, | |
| "accuracy": 0.5445544554455446, | |
| "balanced_accuracy": 0.6624423963133641, | |
| "by_category": { | |
| "abstain": 0.967741935483871, | |
| "answer": 0.35714285714285715 | |
| }, | |
| "leaks": 0, | |
| "mean_tokens": 0.0 | |
| } | |
| }, | |
| "test": { | |
| "n": 156, | |
| "accuracy": 0.5961538461538461, | |
| "balanced_accuracy": 0.6834827388899485, | |
| "by_category": { | |
| "abstain": 0.9183673469387755, | |
| "answer": 0.4485981308411215 | |
| }, | |
| "leaks": 0, | |
| "mean_tokens": 0.0 | |
| }, | |
| "native_vs_cpu": { | |
| "groups": 103, | |
| "delta": 0.36391585760517803, | |
| "interval95": [ | |
| 0.2823624595469255, | |
| 0.44482605177993517 | |
| ] | |
| }, | |
| "seconds_per_request_cpu": 0.02434849743589532, | |
| "rows_sha256": "b3c542237dba50153d88d9358db3c05351bdfbd99fa6b7b87362b9751de5bde3" | |
| } | |