© 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
    3. Gemma-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 8663
    Prev
    Next
    INDEX
    Explanations

    during, assist, most

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

    The neuron selectively activates on emphasis markup—e.g. asterisks (*) or `<em>` tags—used to highlight or italicize words.

    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
     சுமார்
    0.87
     $=
    0.85
     $-
    0.78
     showed
    0.77
     হয়ত
    0.77
     lapply
    0.74
    Perhaps
    0.73
     unui
    0.73
     AKA
    0.72
    ≒
    0.71
    POSITIVE LOGITS
    𝘳
    0.76
    bufs
    0.75
     COMB
    0.74
    ту
    0.72
    oom
    0.71
    ξε
    0.70
    німа
    0.70
    laz
    0.69
    acchati
    0.69
    उट
    0.69
    Activations Density 0.006%

    No Known Activations