© 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. 78709
    Prev
    Next
    INDEX
    Explanations

    multilingual institute names

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

    This neuron fires on mentions of academic or research institutes—especially the token “Instituto” in affiliation/address lines.

    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
    なんだろう
    -0.86
    intellectual
    -0.85
     adquirir
    -0.81
    なと
    -0.80
     Constitucional
    -0.80
    經常
    -0.77
    に行きました
    -0.77
    第二个
    -0.76
    siąż
    -0.75
    íos
    -0.75
    POSITIVE LOGITS
     national
    0.98
     >=",
    0.96
     alles
    0.87
    になる
    0.87
    wiat
    0.85
    accine
    0.84
    illow
    0.84
     de
    0.84
    imarães
    0.83
    ка
    0.81
    Activations Density 0.022%

    No Known Activations