© 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. 24-GEMMASCOPE-2-RES-16K
    7. 9290
    Prev
    Next
    INDEX
    Explanations

    translation or specific programming constructs

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

    The neuron is sensitive to unusually rare or “stand-out” tokens—words or punctuation marks that occur very infrequently in the text.

    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
    ת
    0.91
    ्स
    0.86
     COMPLET
    0.76
     OS
    0.75
     OF
    0.75
     MA
    0.75
     KT
    0.74
     BT
    0.73
    ן
    0.73
    agos
    0.71
    POSITIVE LOGITS
     udział
    1.03
     przede
    0.90
     dzień
    0.88
     poziomie
    0.87
     расходы
    0.85
     tetapi
    0.83
     fueran
    0.83
     tardes
    0.82
     tengan
    0.82
    윅
    0.82
    Activations Density 0.000%

    No Known Activations