© 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. 9930
    Prev
    Next
    INDEX
    Explanations

    various options

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

    The neuron is tuned to spot specific proper‐noun place names (geographical location names).

    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
     أصبح
    0.82
     começar
    0.82
    ُ
    0.81
     iconography
    0.80
     poiché
    0.78
     coworking
    0.78
    눴
    0.77
    <unused2206>
    0.77
     Corbyn
    0.76
     compartió
    0.75
    POSITIVE LOGITS
    肪
    0.84
    E
    0.77
    J
    0.74
    ক্ষন
    0.68
     গুলো
    0.63
    삭제
    0.63
    色々
    0.63
     niest
    0.63
    子供
    0.62
    圾
    0.62
    Activations Density 2.857%

    No Known Activations