© 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. 2487
    Prev
    Next
    INDEX
    Explanations

    The neuron fires strongly on tokens appearing right at the end of each text segment – it’s basically marking the end‐of‐chunk boundary.

    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
    온
    1.01
    드로
    0.88
    드
    0.88
    コンビニ
    0.87
     보면은
    0.82
     latéraux
    0.81
    ポケット
    0.81
    드는
    0.80
     coffers
    0.80
    드가
    0.80
    POSITIVE LOGITS
    ritic
    0.75
     Colin
    0.73
    Unary
    0.69
    ּ
    0.68
    Secondo
    0.67
    chemical
    0.67
    ytale
    0.65
    zym
    0.65
    Colin
    0.65
    stücke
    0.65
    Activations Density 0.000%

    No Known Activations