© 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. 11677
    Prev
    Next
    INDEX
    Explanations

    NO guarantee

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

    The neuron is highly activated by tokens representing numbers or numeric expressions (e.g. digits, decimal values, measurements, citation page/volume numbers).

    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
     caído
    0.82
    적인
    0.78
    ным
    0.78
    진
    0.75
     version
    0.73
     versión
    0.73
     displayName
    0.73
     করো
    0.72
     νό
    0.72
     panda
    0.70
    POSITIVE LOGITS
     อะไร
    0.82
    prache
    0.81
    i
    0.80
    الش
    0.79
    erver
    0.78
    anthropy
    0.78
    biotics
    0.77
    pray
    0.77
    alcool
    0.75
    pecies
    0.75
    Activations Density 0.001%

    No Known Activations