© 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. 65982
    Prev
    Next
    INDEX
    Explanations

    absolutely plus strong positive sentiment

    np_acts-logits-general · gemini-2.5-flash-lite

    absolutely cannot

    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
    し
    1.05
     (
    0.94
    '(
    0.90
     (~
    0.88
    상이
    0.82
     (>
    0.81
    (&
    0.80
    ના
    0.80
     zacz
    0.79
     (`
    0.79
    POSITIVE LOGITS
    is
    1.07
     absolute
    0.98
     for
    0.91
     absolut
    0.81
    ну
    0.78
     absolue
    0.75
    i
    0.72
    ัง
    0.71
     Absolute
    0.69
     moguće
    0.68
    Activations Density 0.022%

    No Known Activations