© 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. 6586
    Prev
    Next
    INDEX
    Explanations

    programatic loops and lists

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

    The neuron specializes in picking out bits of structured metadata or statistical detail—tokens like backslashes in category paths and numeric/statistical values.

    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
     Το
    0.77
     ла
    0.74
     diverge
    0.73
     ка
    0.71
     То
    0.71
     ا
    0.70
     না
    0.70
     주장
    0.69
     investigate
    0.68
     ​​
    0.68
    POSITIVE LOGITS
    ,((
    0.99
    تين
    0.85
    تها
    0.84
    ,(
    0.81
    setAutoDraw
    0.81
    റ്റർ
    0.81
    rosophila
    0.80
    ,-\
    0.80
    ϓ
    0.80
    ᅱ
    0.79
    Activations Density 0.001%

    No Known Activations