© 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. 22-GEMMASCOPE-RES-131K
    7. 35068
    Prev
    Next
    INDEX
    Explanations

    SEC_PLAYER, var lodash, Official Cathedral, 1. SEC prefix checking

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

    The neuron detects uppercase constant identifiers in code (e.g. things like “SEC_PLAYER”).

    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
     continuación
    -1.52
    וּ
    -1.45
     Diputados
    -1.45
     Nedir
    -1.44
     Unies
    -1.31
    prende
    -1.28
    ּוֹ
    -1.24
     бампер
    -1.23
    確か
    -1.22
     Русский
    -1.20
    POSITIVE LOGITS
    櫪
    1.55
     SEVENT
    1.43
    cyklo
    1.43
     gracioso
    1.41
    佷
    1.38
    Fonto
    1.35
    CARACTERIST
    1.35
    Ilustra
    1.33
     kosmety
    1.33
    Probl
    1.31
    Activations Density 0.000%

    No Known Activations