© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 9596
    Prev
    Next
    INDEX
    Explanations

    Buk/Buck and following word

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

    The neuron selectively activates on the substring “Buk,” as in occurrences of the author’s name Bukowski.

    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
     BH
    0.84
     OWN
    0.72
     WORDS
    0.71
     BUFFER
    0.70
     ENTRY
    0.68
     erhältlich
    0.66
    ள
    0.66
    是为了
    0.66
     其实
    0.66
     ബ
    0.65
    POSITIVE LOGITS
    titleTextStyle
    0.70
    mend
    0.66
    proofing
    0.65
    logn
    0.64
    ↵
    0.63
    hood
    0.63
    name
    0.62
    atol
    0.62
    mimeTypes
    0.61
    لات
    0.61
    Activations Density 0.004%

    No Known Activations