© 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. 34-GEMMASCOPE-RES-131K
    7. 58112
    Prev
    Next
    INDEX
    Explanations

    yak, jak, yakima

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

    This neuron is highly sensitive to words beginning with the letter sequence “Yak” (including uppercase variants such as “JAK”).

    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
     spherical
    -0.75
    питы
    -0.74
     voltage
    -0.73
     marine
    -0.73
     resonance
    -0.71
     ТА
    -0.70
     disruptive
    -0.69
     infierno
    -0.69
     prête
    -0.69
    atifying
    -0.69
    POSITIVE LOGITS
     despatch
    0.90
    obs
    0.88
    Yak
    0.85
     Mediator
    0.85
     quat
    0.84
     Stakeholder
    0.84
     proviso
    0.82
    unzel
    0.81
     neutro
    0.81
     labourer
    0.80
    Activations Density 0.013%

    No Known Activations