© 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. 10-GEMMASCOPE-RES-131K
    7. 14980
    Prev
    Next
    INDEX
    Explanations

    Python class initializers

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

    The neuron activates on numeric literals (integers or floats) in the code.

    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
    did
    -1.86
    </h4>
    -1.84
    <em>
    -1.77
     undeniable
    -1.69
     alluring
    -1.65
     decidedly
    -1.64
     unbelievably
    -1.63
    </sup>
    -1.62
     orchestrated
    -1.60
    ついに
    -1.59
    POSITIVE LOGITS
     in
    1.70
     bertahan
    1.66
     on
    1.60
     for
    1.59
    맏
    1.53
    droj
    1.52
    генера
    1.50
    ʡ
    1.48
     as
    1.47
    /*/
    1.47
    Activations Density 0.004%

    No Known Activations