© 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-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 1939
    Prev
    Next
    INDEX
    Explanations

    code comments and people

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

    The neuron fires strongly on semicolon characters, i.e. it detects the presence of “;” punctuation.

    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_31_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
    Estamos
    0.57
     платы
    0.57
    씻
    0.55
    등
    0.54
    Вот
    0.53
    Ж
    0.52
    Finalmente
    0.51
    praw
    0.51
    ഴ
    0.51
    Inicio
    0.50
    POSITIVE LOGITS
     //=>
    0.64
     //@
    0.64
     //[
    0.63
     //.
    0.58
     /**
    0.57
     Concrete
    0.57
     //{
    0.56
     /*
    0.55
     //'
    0.55
    ition
    0.52
    Activations Density 0.000%

    No Known Activations