© 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. 1235
    Prev
    Next
    INDEX
    Explanations

    words like "step" or "less"

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

    The neuron strongly activates on rare or unusual tokens (i.e. long‐tail or out-of-vocabulary tokens).

    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
    Խ
    0.83
     กล่าว
    0.82
     নাহ
    0.79
     اهو
    0.78
     живут
    0.76
     этому
    0.75
     کتاب
    0.74
     نوا
    0.74
     στον
    0.73
    GEBURTSORT
    0.73
    POSITIVE LOGITS
    1
    0.77
    ˉ
    0.77
    5
    0.76
    اری
    0.75
    ри
    0.72
     musculaire
    0.72
    4
    0.69
    т
    0.69
    <0x80>
    0.68
    ể
    0.68
    Activations Density 0.000%

    No Known Activations