© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 9483
    Prev
    Next
    INDEX
    Explanations

    Sudden cardiac death, Payload Assist, Blender Cycles

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

    This neuron strongly activates on the empty‐slice/array notation “[]” in code (i.e. occurrences of “[]” in type declarations).

    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
    s
    -2.44
    紧接着
    -2.20
    l
    -2.17
    Ꭳ
    -2.13
    次は
    -2.05
    r
    -2.03
    U
    -1.91
    そういえば
    -1.89
    来不及
    -1.87
    ﹍﹍
    -1.86
    POSITIVE LOGITS
     as
    2.28
     on
    1.90
     since
    1.88
     what
    1.88
     for
    1.80
     semua
    1.78
     at
    1.73
     even
    1.71
     from
    1.69
     all
    1.68
    Activations Density 0.000%

    No Known Activations