© 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-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 65161
    Prev
    Next
    INDEX
    Explanations

    damage and destruction

    np_acts-logits-general · gemini-2.5-flash-lite

    The neuron chiefly picks out mentions of explicit numerical quantities or statistics in the text.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    üll
    -1.31
     Cedex
    -1.22
    mereka
    -1.21
     Landfill
    -1.17
    ftrag
    -1.13
    嗽
    -1.13
     zand
    -1.11
     shillings
    -1.07
    ulador
    -1.06
    方式
    -1.05
    POSITIVE LOGITS
     damage
    1.79
     niektórych
    1.47
     everything
    1.41
     damages
    1.27
     many
    1.27
     losses
    1.23
     damaged
    1.19
     otwor
    1.19
    ଓ
    1.18
     several
    1.16
    Activations Density 0.046%

    No Known Activations