© 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. 3255
    Prev
    Next
    INDEX
    Explanations

    code structure or syntax

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

    The neuron fires on small, general-purpose “connective” or evaluative words (e.g. “true,” “use,” “means,” “simple,” “great,” “Plate,” “Roof”) that often introduce explanations, evaluations, or results.

    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
    畩
    0.79
     PlayerSelector
    0.77
     Prêmio
    0.76
     São
    0.75
     Vous
    0.74
     Associação
    0.74
     Esses
    0.73
     Serviço
    0.73
    ésil
    0.72
     Hala
    0.71
    POSITIVE LOGITS
    g
    0.75
    bab
    0.74
     synonyms
    0.74
    onders
    0.71
     приви
    0.71
     খ্যা
    0.71
    ters
    0.70
     litter
    0.69
     encrypt
    0.68
    iva
    0.68
    Activations Density 0.001%

    No Known Activations