© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-27B-IT
    5. 41-GEMMASCOPE-2-RES-262K
    7. 26800
    Prev
    Next
    INDEX
    Explanations

    malicious intent or actions

    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
    Negative Logits
     angustato
    0.43
    蘞
    0.42
    蒨
    0.41
    脨
    0.41
     Disbursements
    0.41
    ncnc
    0.41
    粓
    0.41
    विच
    0.39
    惻
    0.39
    findContacts
    0.38
    POSITIVE LOGITS
     abuse
    1.05
     unscrupulous
    1.04
     loopholes
    0.95
     manipulation
    0.91
     malicious
    0.91
     abuses
    0.90
     loophole
    0.90
     abusers
    0.89
     manipulative
    0.88
     fraudsters
    0.86
    Activations Density 0.002%

    No Known Activations