© 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. 31-GEMMASCOPE-2-RES-65K
    7. 62195
    Prev
    Next
    INDEX
    Explanations

    don't bother trying

    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_31_width_65k_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
    0.48
     counselors
    0.46
     accolades
    0.45
     are
    0.45
     envisioned
    0.43
     precluded
    0.43
     A
    0.43
     these
    0.43
     </
    0.42
     whose
    0.42
    POSITIVE LOGITS
    त
    0.63
    Vì
    0.50
     ಜೀವನ
    0.49
     ძალიან
    0.48
     공부해
    0.48
    ށ
    0.48
     الدفاع
    0.48
    Cómo
    0.47
    رام
    0.47
    ównie
    0.47
    Activations Density 0.010%

    No Known Activations