© 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. 3687
    Prev
    Next
    INDEX
    Explanations

    role descriptions

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

    The neuron fires strongly on blocks of text that read like image or figure captions—i.e. concise visual‐scene descriptions.

    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
    &
    0.84
     §§
    0.75
     &
    0.72
    >
    0.71
     update
    0.71
     FAQs
    0.71
     prove
    0.70
    $
    0.69
     Update
    0.69
     FAQ
    0.69
    POSITIVE LOGITS
     copyspace
    1.18
     stockfoto
    1.05
     фото
    1.02
     художника
    1.00
     escenas
    0.99
     illustrazione
    0.97
     фотографии
    0.95
     fotografía
    0.95
     fotografías
    0.95
    美しい
    0.95
    Activations Density 0.047%

    No Known Activations