© 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-TRANSCODER-262K
    7. 76397
    Prev
    Next
    INDEX
    Explanations

    defense mechanism and avoidance

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-27b-it/transcoder_all/layer_31_width_262k_l0_small_affine
    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
    သင့်
    0.39
     تربیت
    0.38
     組
    0.37
     സന്ത
    0.36
    uddho
    0.36
     sóng
    0.36
    simmons
    0.35
    DBO
    0.34
     enlightened
    0.34
    erreur
    0.34
    POSITIVE LOGITS
     deflect
    0.48
     compensatory
    0.45
     defensive
    0.44
    回避
    0.44
     compensate
    0.44
     compensates
    0.44
    逃
    0.44
     avoidance
    0.43
    を防
    0.43
    compens
    0.41
    Activations Density 0.148%

    No Known Activations