© 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. 6487
    Prev
    Next
    INDEX
    Explanations

    followed by common words

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

    The neuron responds strongly to relatively uncommon, low‐frequency words (i.e. uncommon vocabulary items).

    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
    穣
    0.73
    lib
    0.70
    don
    0.68
    pong
    0.66
    cultural
    0.65
    ことができる
    0.65
     බ
    0.64
     i
    0.63
     poch
    0.63
    icon
    0.63
    POSITIVE LOGITS
    ॉक्स
    0.92
     waveguides
    0.91
     гуляць
    0.90
     также
    0.87
    ссия
    0.87
    ваты
    0.86
     همچنین
    0.84
     tiež
    0.84
     terceiro
    0.83
    льный
    0.82
    Activations Density 0.000%

    No Known Activations