© 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. 12579
    Prev
    Next
    INDEX
    Explanations

    technical merits, conversion, intestines

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

    The neuron is a high-activation detector for very frequent/common tokens—essentially “stop words” or other high-frequency words (e.g. to, the, in, is, over).

    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
    స
    0.75
    satz
    0.71
    کنندگان
    0.71
    a
    0.71
    tion
    0.68
    sas
    0.68
    Q
    0.68
    WWW
    0.66
    만을
    0.66
    srt
    0.65
    POSITIVE LOGITS
    տ
    0.84
     seductive
    0.80
    hiya
    0.78
    филлер
    0.78
     perder
    0.76
     unparalleled
    0.76
    uy
    0.75
    tempCard
    0.74
     espécie
    0.73
    barer
    0.73
    Activations Density 0.000%

    No Known Activations