© 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. 2926
    Prev
    Next
    INDEX
    Explanations

    citations [number]

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

    The neuron detects parenthesized numeric citations or reference markers, especially floating‐point numbers in parentheses.

    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
    <0xBE>
    0.72
     paradigma
    0.72
    opting
    0.70
    formatics
    0.70
     लिखेंगे
    0.68
    ＜
    0.68
     pessimistic
    0.66
    ∶
    0.66
    <0xB7>
    0.65
    を用いる
    0.65
    POSITIVE LOGITS
    Greeting
    0.69
    것도
    0.69
    LetterIndex
    0.69
    껏
    0.68
    सिर
    0.68
     पनि
    0.67
    ChangeString
    0.67
    Geschichte
    0.66
     chiaro
    0.66
    Passengers
    0.66
    Activations Density 0.000%

    No Known Activations