© 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-TRANSCODER-262K
    7. 28187
    Prev
    Next
    INDEX
    Explanations

    undermine, erosion, sabotage

    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_40_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
     dich
    0.42
    Slot
    0.37
    丟
    0.37
     తమిళ
    0.37
     Slot
    0.36
     spic
    0.35
    Closure
    0.34
     mardi
    0.34
    礻
    0.34
     viewController
    0.34
    POSITIVE LOGITS
     undermine
    1.91
     undermined
    1.90
     undermining
    1.88
     undercut
    1.73
     undermines
    1.69
     underm
    1.05
     eroding
    0.97
     erode
    0.96
    Und
    0.88
     Und
    0.88
    Activations Density 0.103%

    No Known Activations