© 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. 31-GEMMASCOPE-2-RES-16K
    7. 2897
    Prev
    Next
    INDEX
    Explanations

    SAA, plain text, sauté

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

    The neuron specifically fires on the token “Sayoko,” i.e. it detects occurrences of that proper name.

    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
    pretty
    0.54
     informiert
    0.53
    ちょっと
    0.52
    鏨
    0.52
    セミ
    0.51
    সির
    0.51
    প্রধান
    0.51
    similar
    0.50
    Similar
    0.50
    
    0.50
    POSITIVE LOGITS
     rule
    0.55
     NORTH
    0.54
    aved
    0.54
     canv
    0.53
     ساف
    0.52
    erkraut
    0.50
     माधव
    0.50
     attrition
    0.49
    ()'
    0.49
    тта
    0.49
    Activations Density 0.159%

    No Known Activations