© 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. 24-GEMMASCOPE-2-RES-16K
    7. 2099
    Prev
    Next
    INDEX
    Explanations

    phrases ending a turn, numbers, or specific words

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

    The neuron fires strongly on numeric tokens formatted as decimal numbers.

    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
    hernalia
    0.76
     conç
    0.73
     ἢ
    0.71
     Göttingen
    0.70
     Pé
    0.69
     functor
    0.69
    ística
    0.68
     நிறுவ
    0.68
     hinders
    0.68
     rétréc
    0.68
    POSITIVE LOGITS
    خ
    0.86
    1
    0.84
    מ
    0.83
    נ
    0.82
    re
    0.82
    4
    0.81
    9
    0.80
    ל
    0.80
    3
    0.79
    6
    0.77
    Activations Density 0.001%

    No Known Activations