© 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-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 100346
    Prev
    Next
    INDEX
    Explanations

    leaving a state or consequence

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

    The neuron spots mentions of quantitative “left <numerical amount>” phrases (e.g. “left 320,” “left 316”) that report counts, especially casualty or population figures.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     to
    -2.28
     the
    -2.19
     of
    -1.54
     about
    -1.53
     is
    -1.49
    u
    -1.49
    gulier
    -1.46
    </i>
    -1.46
    on
    -1.45
    .”
    -1.45
    POSITIVE LOGITS
     niñas
    1.84
     trajes
    1.82
     joyas
    1.78
    諛
    1.77
     élégante
    1.77
     ecuator
    1.75
     faldas
    1.73
    发生的
    1.70
     Наталья
    1.70
     decoracion
    1.68
    Activations Density 0.012%

    No Known Activations