Text Classification
Transformers
Safetensors
English
Korean
Japanese
qev_decision_merged
feature-extraction
qev
typed-decision
custom-code
bf16
custom_code
Instructions to use querypieai/QEV-9B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use querypieai/QEV-9B with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-classification", model="querypieai/QEV-9B", trust_remote_code=True)# pip install -U transformers accelerate # Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("querypieai/QEV-9B", trust_remote_code=True, device_map="auto") - Notebooks
- Google Colab
- Kaggle
File size: 3,329 Bytes
96adfbe | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 | {
"datasets_distributed": false,
"processed_for_qev_comparative_evaluation": true,
"official_unmodified_benchmark_scores": false,
"public_comparison_context": {"state": 384, "per_question_row": 1024, "packed": 2048, "truncate": false, "selection": "fixed successful item-ID intersection across compared original models"},
"datasets": {
"mmluprox-lite": {
"source_url": "https://huggingface.co/datasets/li-lab/MMLU-ProX-Lite",
"source_revision_recorded_by_manifest": "e82aafb9460529687d3c7e51b401d8dd1dd309dd",
"source_license": "MIT",
"split": "test",
"languages": 29,
"source_questions": 17052,
"reported_questions": 16384,
"processing": "question to state; valid options to choice criteria; cot_content excluded; blank/N/A choices removed; original question/answer and option order retained and validated; fixed common item-ID subset",
"processed_sha256": "46b4e624eba8f41d002ea215c5ff03661fbfc5d31837e7833fbd574f6976b348"
},
"belebele": {
"source_url": "https://huggingface.co/datasets/facebook/belebele",
"source_revision_recorded_by_manifest": "7899cdfa4e1e0d733fd77c848e2c273cb1d32be2",
"source_license": "CC-BY-SA-4.0",
"split": "test",
"languages": 29,
"source_questions": 26100,
"reported_questions": 24622,
"processing": "flores_passage to state; original question to instructions; four mc_answers to choice criteria; correct_answer_num mapped to label; fixed common item-ID subset; processed passages not redistributed",
"processed_sha256": "fcae61bceeb6953c62b163dd495614bd2a28e04871f82b9a05a14ed86604e8e8"
},
"private-typed-decision": {
"source": "QueryPie project private-general-v1",
"public_download": false,
"label_generation": "deterministic code/rule solver; seed 20261006",
"translation": "assistant-authored deterministic EN to KO/JA template localization; labels, numbers and option keys preserved; automatic integrity checks; no native-speaker review",
"source_states": 450,
"source_questions": 900,
"reported_states": 270,
"reported_questions": 540,
"questions_per_language": {"en": 180, "ko": 180, "ja": 180},
"processing": "choice/noul/score questions; exclude the entire state if q1/q2 type, instructions and criteria are identical; identical selected IDs used for every model",
"processed_sha256": "5c061e647377b909fde9368de002944ba1ca7e9bb0963facc5ad8e9482b1b5d2"
},
"dlp": {
"source": "QueryPie internal held-out multilingual DLP test corpus",
"public_download": false,
"documents": 3167,
"positive": 1281,
"negative": 1886,
"documents_by_language": {"en": 1065, "ja": 1038, "ko": 1064},
"processing": "original text/IDs/labels retained; document to state; English noul true/false instruction; any ground-truth finding means DLP positive; no finding-span/count/type extraction metric",
"threshold": 0.5,
"processed_sha256": "77cd5fbd3ef74695912f008c0838dd5fd66462278167259fc4e73a182c3a2e97"
}
},
"composite_definition": "equal-weight mean of question-level accuracies of MMLU-ProX-Lite, Belebele and private Typed-Decision",
"new_inference_for_card_update": false,
"raw_data_and_item_level_predictions_included": false
}
|