© 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. 14758
    Prev
    Next
    INDEX
    Explanations

    function names and variables

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

    This neuron specifically detects occurrences of the letter “g” used as a mathematical variable (often with subscripts or numeric annotations) in formulaic or technical expressions.

    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
    ୪
    -2.09
    硖
    -2.05
    Jahre
    -1.89
    tieve
    -1.79
    quelize
    -1.73
    闿
    -1.73
    ọi
    -1.72
    pantalon
    -1.71
    élène
    -1.68
    赝
    -1.67
    POSITIVE LOGITS
    8
    2.58
    9
    2.50
    >
    2.33
    ]
    2.30
    C
    2.23
    *
    2.23
    7
    2.22
    N
    2.16
    D
    2.13
    6
    2.13
    Activations Density 0.025%

    No Known Activations