© 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. 8045
    Prev
    Next
    INDEX
    Explanations

    autotrophically, medical journal, per year

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

    The neuron responds to uncommon, content-bearing words (i.e. rare or distinctive tokens).

    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
    衹
    0.73
    是
    0.71
    টা
    0.66
    作者
    0.66
     extrémités
    0.66
     REI
    0.66
    ট
    0.66
    呐
    0.64
    你是
    0.64
    演員
    0.63
    POSITIVE LOGITS
     สาว
    0.85
     kıl
    0.82
    ل
    0.79
    inson
    0.77
    ambique
    0.73
     เนื่อง
    0.73
     סי
    0.72
     Thunderbird
    0.71
     เบ
    0.70
     รูป
    0.70
    Activations Density 0.000%

    No Known Activations