© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 2575
    Prev
    Next
    INDEX
    Explanations

    contextual descriptions of tasks

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

    The neuron responds primarily to unusually long tokens—i.e. words or names that span many characters.

    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
    ずつ
    0.81
     Treatments
    0.78
     Didn
    0.75
    ḕ
    0.73
    があれば
    0.72
    estruct
    0.70
    imetable
    0.70
     모습을
    0.70
     intención
    0.68
     aparición
    0.68
    POSITIVE LOGITS
    i
    0.99
    n
    0.90
    se
    0.88
    н
    0.87
    the
    0.83
    드
    0.82
    mc
    0.79
    mac
    0.79
    seq
    0.79
    motif
    0.78
    Activations Density 0.000%

    No Known Activations