© 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
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 128668
    Prev
    Next
    INDEX
    Explanations

    attack or attacked

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

    The neuron detects occurrences of the word “attack” (in any inflected form such as attack, attacked, attacking).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
    It
    -2.38
    })}
    -2.06
    …..
    -1.99
    Y
    -1.96
    and
    -1.93
    p
    -1.86
    ’
    -1.86
    on
    -1.85
    I
    -1.84
    H
    -1.84
    POSITIVE LOGITS
     escuchadas
    2.20
    蟜
    2.13
     heridos
    2.03
     ogrom
    2.03
     choroby
    2.02
     testigos
    1.98
     Behaup
    1.95
     obrazy
    1.94
     velmi
    1.93
     bedenken
    1.93
    Activations Density 0.029%

    No Known Activations