© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 15013
    Prev
    Next
    INDEX
    Explanations

    time zone

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

    The neuron responds to alphanumeric technical identifiers (like file‐format and device names that mix letters and digits, e.g. “PS3,” “MP3,” “MP4,” “PCM,” 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
    Imagenes
    0.84
    ে
    0.84
     Гульнявыя
    0.83
    Asimismo
    0.83
     Спорттук
    0.83
     Zoology
    0.81
     Idani
    0.80
     Aktual
    0.80
    Leggi
    0.79
    Azer
    0.77
    POSITIVE LOGITS
    /
    0.91
    +
    0.91
     it
    0.76
     or
    0.75
    ities
    0.74
    ikes
    0.74
    ite
    0.73
     jako
    0.72
    вает
    0.71
    oru
    0.71
    Activations Density 2.257%

    No Known Activations