© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 4314
    Prev
    Next
    INDEX
    Explanations

    commands after backticks

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

    The neuron detects bibliographic citation markers, especially bracketed reference tags (e.g. “[…]”) and related reference tokens like “Internet.”

    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
    I
    0.80
    in
    0.79
     d
    0.77
    a
    0.77
     quarries
    0.76
    y
    0.75
     ఇ
    0.74
    ли
    0.73
    en
    0.73
    ی
    0.71
    POSITIVE LOGITS
    喈
    0.89
    @@]
    0.81
    േഷ്
    0.80
    ㄽ
    0.78
    getCharAt
    0.77
    ๙
    0.77
     titulado
    0.76
    ྞ
    0.75
    lycer
    0.75
    ].”
    0.75
    Activations Density 0.001%

    No Known Activations