© 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. 5125
    Prev
    Next
    INDEX
    Explanations

    left with A and B

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

    The neuron spikes on mentions of formal organization or institution names (e.g. universities, companies, commissions).

    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
    дачи
    0.86
    ateurs
    0.80
    ía
    0.79
    كن
    0.78
     Assets
    0.76
    თვის
    0.75
     ﺍ
    0.74
     último
    0.73
     Education
    0.71
     čin
    0.71
    POSITIVE LOGITS
    gtk
    0.83
     unenforceable
    0.81
     revoc
    0.75
    뚠
    0.75
     vectorize
    0.74
    裒
    0.74
     concatenate
    0.72
     mitigate
    0.71
    ECUTIVE
    0.71
    Kp
    0.71
    Activations Density 0.000%

    No Known Activations