Instructions to use SkillScreen/SkillScreen-2B-SD with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use SkillScreen/SkillScreen-2B-SD with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-classification", model="SkillScreen/SkillScreen-2B-SD")# pip install -U transformers accelerate # Load model directly from transformers import AutoProcessor, AutoModelForSequenceClassification processor = AutoProcessor.from_pretrained("SkillScreen/SkillScreen-2B-SD") model = AutoModelForSequenceClassification.from_pretrained("SkillScreen/SkillScreen-2B-SD", device_map="auto") - Notebooks
- Google Colab
- Kaggle
|
Download THIRD_PARTY_NOTICES.md from SkillScreen/SkillScreen-2B-SD: direct link, hf CLI and curl.
- Browser
- Download file 984 Bytes
-
https://huggingface.co/SkillScreen/SkillScreen-2B-SD/resolve/main/THIRD_PARTY_NOTICES.md
- Command line
-
hf download hf://SkillScreen/SkillScreen-2B-SD/THIRD_PARTY_NOTICES.md
-
curl -L -o THIRD_PARTY_NOTICES.md https://huggingface.co/SkillScreen/SkillScreen-2B-SD/resolve/main/THIRD_PARTY_NOTICES.md
984 Bytes
| # Third-party notices | |
| SkillScreen-authored code, configuration, code examples, and our contributions to model weights and adapters are licensed under Apache-2.0 (LICENSE). Documentation, citation metadata, and released provenance/evaluation reports are licensed under CC BY 4.0 (LICENSE-CC-BY-4.0). See README.md for scope. Upstream rights remain applicable. | |
| Base model [OpenJev](https://huggingface.co/AlexWortega/openjev/tree/a20448012c213128955ca0c693e7c943865cab77) is designated MIT by its model card. It derives from [Qwen3.5-2B](https://huggingface.co/Qwen/Qwen3.5-2B), licensed Apache-2.0. Upstream notices are included in LICENSE-OPENJEV and LICENSE-QWEN. | |
| Training benchmark [MaliciousSkillBench](https://huggingface.co/datasets/ProtectSkills/MaliciousSkillBench/tree/d4b42ce5766a6e0359c987cf59c1007cb3795a90), Yue Wang et al. Benchmark metadata and splits are CC BY 4.0. Skill texts and packages retain upstream terms. No training texts or packages are distributed here. | |