© 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. 41-GEMMASCOPE-2-RES-262K
    7. 30544
    Prev
    Next
    INDEX
    Explanations

    shift focus away

    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_all/layer_41_width_262k_l0_big
    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
    တင်
    0.43
     روح
    0.39
     Consistency
    0.39
     આ
    0.39
    anner
    0.38
     Цвет
    0.38
     Selanjutnya
    0.38
    ρχ
    0.37
     regularity
    0.37
    绎
    0.37
    POSITIVE LOGITS
     away
    1.00
    Away
    0.87
     AWAY
    0.80
     hacia
    0.79
    away
    0.77
     Away
    0.77
     towards
    0.74
     toward
    0.74
    towards
    0.71
     elsewhere
    0.63
    Activations Density 0.000%

    No Known Activations