© 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. 1701
    Prev
    Next
    INDEX
    Explanations

    The neuron is simply tracking how early a token appears in the input (strongly activating at the start and steadily decaying over the course of the document).

    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
     podía
    1.01
     quería
    0.98
    avacak
    0.93
    ayvachi
    0.92
     estratégico
    0.92
    덫
    0.90
     bloquear
    0.90
     автомобиля
    0.89
     reiniciar
    0.89
     Biomed
    0.89
    POSITIVE LOGITS
    1.06
    L
    0.80
    Vi
    0.77
    Summary
    0.77
    NA
    0.76
    Mens
    0.74
    ς
    0.74
    Ne
    0.73
    Previous
    0.73
    Ladies
    0.72
    Activations Density 0.013%

    No Known Activations