© 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. 12872
    Prev
    Next
    INDEX
    Explanations

    Okay, let's break down

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

    The neuron fires strongly on punctuation marks (especially commas and question‐mark tokens), essentially marking clause or phrase separators.

    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
    ,
    0.88
     —
    0.73
    ².
    0.68
     fascia
    0.65
    w
    0.63
    i
    0.62
    >();
    0.61
    f
    0.60
     /
    0.59
     $+$
    0.59
    POSITIVE LOGITS
    甛
    0.92
     odnosno
    0.89
     yea
    0.86
     iniziamo
    0.79
    如果在
    0.78
    如果你
    0.77
    andeep
    0.77
     inizia
    0.77
     ilość
    0.77
     Jeżeli
    0.75
    Activations Density 0.165%

    No Known Activations