© 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. 22105
    Prev
    Next
    INDEX
    Explanations

    machine names and brands

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

    The neuron fires on the names of concrete objects or products (e.g. “gummy bear(s),” “excavator(s),” “backhoes,” “Toshiba”).

    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
    ata
    -1.78
    ite
    -1.73
    доволь
    -1.64
     על
    -1.63
    issime
    -1.59
    がありました
    -1.54
     legyen
    -1.53
    ですが
    -1.52
    ’,
    -1.52
    おき
    -1.52
    POSITIVE LOGITS
     preparados
    2.16
     conectado
    2.00
     atentos
    1.96
    Bety
    1.94
     destinados
    1.84
    风险
    1.84
    幛
    1.83
    下面的
    1.78
     químico
    1.77
    ープン
    1.73
    Activations Density 0.014%

    No Known Activations