© 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. 12681
    Prev
    Next
    INDEX
    Explanations

    VRAM, versions, pain, target

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

    This neuron fires most strongly on tokens that single out or refer to a target group or category (e.g. “those,” “patients,” “website visitors,” etc.).

    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
    olon
    0.75
    $
    0.75
     kaž
    0.73
    Phi
    0.71
     Gesture
    0.70
     eyepiece
    0.70
    StudentNo
    0.68
     مشاهد
    0.68
    )]$
    0.68
     orci
    0.67
    POSITIVE LOGITS
    스의
    0.79
    आरओ
    0.78
    ซ์
    0.75
    প্তিতে
    0.75
    enangkan
    0.73
    शियन
    0.73
    浻
    0.73
    ടതി
    0.72
    ھیں
    0.71
    द्री
    0.71
    Activations Density 0.000%

    No Known Activations