© 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. 22-GEMMASCOPE-2-TRANSCODER-262K
    7. 41173
    Prev
    Next
    INDEX
    Explanations

    bypassing safety filters

    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_22_width_262k_l0_small_affine
    Prompts (Dashboard)
    238,145 prompts, 512 tokens each
    Dataset (Dashboard)
    lmsys + oasst1
    No Configuration Found
    Embeds
    IFrame
    Link
    Negative Logits
     blinking
    0.33
     terrible
    0.33
     scintillation
    0.32
     member
    0.32
     drunken
    0.31
     freelancer
    0.31
     suction
    0.30
     درخواست
    0.30
     nebo
    0.30
    &-
    0.30
    POSITIVE LOGITS
    ridged
    0.41
    affirming
    0.38
    competitive
    0.37
    openjdk
    0.37
    যু
    0.36
    claiming
    0.35
    spired
    0.35
    kében
    0.35
    produce
    0.34
    crete
    0.33
    Activations Density 0.000%

    No Known Activations