© 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. 31-GEMMASCOPE-2-RES-16K
    7. 8395
    Prev
    Next
    INDEX
    Explanations

    states, or conditions

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

    This neuron isn’t picking out a semantic pattern in the text but rather fires strongly on tokens that occur at a particular absolute position in the input (around the 2,700–2,800-token mark).

    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_31_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
    اندی
    0.77
    ité
    0.68
    %%%%%%%%%%%%%
    0.67
    nosti
    0.65
    price
    0.63
    alities
    0.63
    renders
    0.63
    neſs
    0.63
    an
    0.61
    al
    0.61
    POSITIVE LOGITS
     Wii
    0.82
    층
    0.82
    ло
    0.80
     думаю
    0.79
    상은
    0.78
    鯢
    0.77
     workbook
    0.76
    긴
    0.76
    銠
    0.75
     floppy
    0.72
    Activations Density 0.000%

    No Known Activations