© 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. 4015
    Prev
    Next
    INDEX
    Explanations

    branches prefixed

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

    The neuron fires strongly on the very first significant word or token at the start of a new passage or section (e.g. section headers or opening words of document chunks).

    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
    bine
    0.90
    al
    0.87
    ن
    0.81
    irmer
    0.80
    এরপর
    0.73
     Таким
    0.71
    د
    0.71
     Skeleton
    0.70
    alions
    0.70
    isso
    0.69
    POSITIVE LOGITS
    tal
    0.85
    tır
    0.82
    です
    0.80
     wartość
    0.76
     nilai
    0.75
    té
    0.75
     tém
    0.75
     value
    0.74
     chóng
    0.73
    ▿
    0.73
    Activations Density 0.000%

    No Known Activations