© 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. 72515
    Prev
    Next
    INDEX
    Explanations

    Standard dose

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

    The neuron fires on tokens that are pure hexadecimal values (e.g. two-digit hex byte codes or standalone hex digits).

    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
     this
    -1.16
     after
    -1.02
     to
    -1.02
    ênes
    -0.94
    وی
    -0.92
     anderem
    -0.91
    ografi
    -0.90
     itself
    -0.90
    jdbc
    -0.88
    Index
    -0.85
    POSITIVE LOGITS
     behövs
    0.96
     geblieben
    0.91
     라이
    0.90
    曲は
    0.90
    lition
    0.89
     ܙ
    0.88
     samtidigt
    0.87
     pienso
    0.86
    ୀ
    0.85
     проходит
    0.85
    Activations Density 0.000%

    No Known Activations