© 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. 27143
    Prev
    Next
    INDEX
    Explanations

    players manage; implied context; grammatical gender; driven off

    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
     dominated
    0.41
     well
    0.39
     grouping
    0.38
     called
    0.37
    :
    0.37
    <0xB7>
    0.37
    ogh
    0.36
    _
    0.36
    di
    0.35
    oph
    0.35
    POSITIVE LOGITS
    阿里云
    0.46
    愊
    0.46
    한다고
    0.43
    ܀
    0.43
     وعند
    0.42
     வடக்கு
    0.42
    ܂
    0.41
     دوسری
    0.41
    stanbul
    0.41
     Sereth
    0.41
    Activations Density 0.000%

    No Known Activations