© 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-IT
    5. 40-GEMMASCOPE-2-RES-262K
    7. 131889
    Prev
    Next
    INDEX
    Explanations

    safe and warm

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-27b-it/resid_post/layer_40_width_262k_l0_medium
    Prompts (Dashboard)
    238,145 prompts, 512 tokens each
    Dataset (Dashboard)
    lmsys + oasst1
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    rebbero
    0.45
    onent
    0.43
    ContextHeader
    0.42
     देशभर
    0.41
    衽
    0.41
    archiviato
    0.41
    ள்ளார்
    0.40
     thicker
    0.40
    DialogWhen
    0.39
     schal
    0.39
    POSITIVE LOGITS
    MARK
    0.43
     solving
    0.40
     Mark
    0.39
     finding
    0.39
     INF
    0.39
     Homework
    0.39
     solve
    0.38
    공
    0.38
     Ad
    0.38
     suspend
    0.37
    Activations Density 0.000%

    No Known Activations