© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-27B
    5. 31-GEMMASCOPE-2-RES-262K
    7. 134844
    Prev
    Next
    INDEX
    Explanations

    threatening concepts

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-27b-pt/resid_post/layer_31_width_262k_l0_medium
    Prompts (Dashboard)
    392,802 prompts, 256 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    綘
    0.47
     подключения
    0.46
     Př
    0.46
    颜值
    0.45
     Helped
    0.44
     만들어
    0.43
    劳动
    0.43
    電動
    0.43
     Cara
    0.43
     비밀
    0.43
    POSITIVE LOGITS
     threatening
    0.49
    threatening
    0.49
    dent
    0.48
    n
    0.47
    str
    0.47
     degraded
    0.46
    communicate
    0.46
    threat
    0.46
    coordinate
    0.46
     correspondingly
    0.46
    Activations Density 0.002%

    No Known Activations