© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 13049
    Prev
    Next
    INDEX
    Explanations

    The neuron lights up on unusually long or specific alphanumeric substrings—e.g. email‐address usernames, procedural names, or numeric codes—rather than ordinary words.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-12b-pt/resid_post/layer_24_width_16k_l0_medium
    Prompts (Dashboard)
    392,802 prompts, 256 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    .
    0.96
    ,
    0.93
    :
    0.86
    !
    0.84
    ;
    0.71
     and
    0.68
     counties
    0.65
     USA
    0.64
    m
    0.64
    -
    0.64
    POSITIVE LOGITS
     पहली
    0.87
     بسیار
    0.86
     சமீப
    0.86
     recientemente
    0.85
     wanneer
    0.84
     요즘
    0.84
     در
    0.83
    Pada
    0.83
    難しい
    0.83
     недостаточно
    0.82
    Activations Density 0.433%

    No Known Activations