© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Gemma-3-12B
    3. 24-GEMMASCOPE-2-RES-16K
    4. 10332
    Prev
    Next
    INDEX
    Explanations

    formatting and special characters

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

    This neuron strongly activates on code‐comment text (especially inline comments marked with “//” or containing TODO/FIXME/NOTE annotations).

    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
     austenite
    0.66
     lyophil
    0.66
     lipoproteins
    0.64
     a
    0.63
     antioxid
    0.61
     antidepressants
    0.59
    ürich
    0.59
    <strong>
    0.59
     decolor
    0.58
    évolution
    0.57
    POSITIVE LOGITS
     Например
    0.94
     FIXME
    0.87
     这里
    0.83
     TODO
    0.80
     Пример
    0.77
     Örneğin
    0.77
     Misalnya
    0.77
     requires
    0.75
     Here
    0.74
     Requires
    0.73
    Activations Density 0.009%

    No Known Activations