© 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. 11444
    Prev
    Next
    INDEX
    Explanations

    the newborn

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

    The neuron fires on agent-type nouns referring to people or groups of people (e.g. “mariners,” “terrorist”).

    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
    ্দ্র
    0.66
     enrolled
    0.62
    naire
    0.60
    <0x81>
    0.59
     recurrence
    0.58
    OUSE
    0.57
     entirety
    0.57
     topologically
    0.57
    urrences
    0.57
    chen
    0.56
    POSITIVE LOGITS
     далі
    0.86
    回报
    0.85
    矫
    0.80
     Tät
    0.79
     rượu
    0.79
    Đi
    0.79
    rpm
    0.78
    Quand
    0.77
    RPM
    0.77
     antipode
    0.77
    Activations Density 0.001%

    No Known Activations