© 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. 33440
    Prev
    Next
    INDEX
    Explanations

    array followed by A or related elements

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

    The neuron flags numeric tokens—literal numbers—especially within code or technical descriptions.

    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
    </td>
    -1.27
    Ecotoxicity
    -1.04
    )';
    -0.98
    >?
    -0.96
    ;&
    -0.95
     or
    -0.94
    }>
    -0.93
    tomime
    -0.92
    plication
    -0.91
     Tieren
    -0.91
    POSITIVE LOGITS
     array
    1.14
     μας
    1.11
    ция
    1.05
    andidaten
    1.02
    べた
    1.01
    Klik
    0.99
     spē
    0.98
     DETECTION
    0.98
    ție
    0.97
     baterie
    0.96
    Activations Density 0.033%

    No Known Activations