© 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-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 32744
    Prev
    Next
    INDEX
    Explanations

    per se

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

    The neuron is tuned to spot the Latin phrase “per se.”

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
     byste
    -0.83
     člověk
    -0.79
    serie
    -0.76
     produzir
    -0.75
    नीय
    -0.75
     differenti
    -0.75
     tedy
    -0.74
     Attachments
    -0.74
    Meteor
    -0.73
     ilman
    -0.73
    POSITIVE LOGITS
     se
    2.39
     say
    1.58
     Se
    1.56
    Se
    1.29
     see
    1.29
     се
    1.25
    say
    1.21
     says
    1.09
    see
    1.03
     sich
    1.00
    Activations Density 0.033%

    No Known Activations