© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 12530
    Prev
    Next
    INDEX
    Explanations

    Do this, sequence, relying on, mastered, relatively

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

    This neuron is highly sensitive to numeric tokens and figures (dates, times, measurements, and other 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
    olution
    0.84
     kreeg
    0.80
     generan
    0.80
     olv
    0.80
    ”。
    0.79
    ”।
    0.79
    ha
    0.78
     ৷
    0.78
    ondi
    0.78
    惟
    0.78
    POSITIVE LOGITS
    ان
    0.79
    Caracter
    0.76
    Celtic
    0.73
    رشف
    0.72
    cystic
    0.71
    zelfde
    0.71
    об
    0.70
    Caucasian
    0.69
    grapes
    0.69
     auditions
    0.68
    Activations Density 0.001%

    No Known Activations