© 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. 11544
    Prev
    Next
    INDEX
    Explanations

    choosing platform or selecting voice

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

    The neuron fires strongly on the first meaningful word of each new text segment (i.e. the initial content token at the start of a sentence or paragraph).

    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
    स
    1.01
    s
    0.90
    ات
    0.87
    Istituto
    0.86
    ्स
    0.86
    У
    0.83
    ાન
    0.75
    Cookies
    0.75
    Χ
    0.75
     Istituto
    0.75
    POSITIVE LOGITS
    чное
    0.88
     зарабо
    0.86
    сы
    0.84
     arrec
    0.82
    aggregation
    0.79
    fromString
    0.79
    еты
    0.78
     maim
    0.76
     второй
    0.75
     configurational
    0.75
    Activations Density 0.000%

    No Known Activations