© 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. 6322
    Prev
    Next
    INDEX
    Explanations

    machine parts and devices

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

    The neuron is primarily detecting numeric tokens or numerical expressions (e.g. digits, measurements, figures).

    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
     specializes
    0.86
    வுக
    0.68
     specialises
    0.67
     тре
    0.67
     מפ
    0.66
     chuyên
    0.66
     specialize
    0.66
    ו
    0.66
     Specialists
    0.65
    haf
    0.65
    POSITIVE LOGITS
    嚥
    0.84
    ப்பையும்
    0.79
    寒
    0.75
     hInstance
    0.75
     Smit
    0.75
     kneel
    0.75
    விலான
    0.75
    અમ
    0.75
     vines
    0.73
    峯
    0.73
    Activations Density 0.000%

    No Known Activations