© 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
    2. Gemma-3-12B
    3. 24-GEMMASCOPE-2-RES-16K
    4. 11119
    Prev
    Next
    INDEX
    Explanations

    age problems

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

    The neuron fires on calls-to-action or promotional/instructional content (e.g. recipe steps, sales pitches, “take care of your skin,” “contact us,” ads).

    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
     of
    0.86
    ————
    0.83
     $\
    0.78
     jaar
    0.78
    um
    0.74
     to
    0.74
    /
    0.71
     snd
    0.70
     in
    0.70
    ElementChild
    0.69
    POSITIVE LOGITS
    𝙀
    0.83
    खाव
    0.77
    を受けた
    0.76
    ições
    0.76
    𝗴
    0.75
     Yesus
    0.75
    極
    0.75
    🫦
    0.75
     يتم
    0.73
    therners
    0.73
    Activations Density 0.000%

    No Known Activations