© 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. 113757
    Prev
    Next
    INDEX
    Explanations

    cherished people and things

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

    This neuron specifically detects the appearance of the word “beloved.”

    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
     -
    -2.98
     perfecta
    -2.44
     We
    -2.41
    </u>
    -2.33
    踉
    -2.28
     amables
    -2.20
    厭
    -2.17
    ar
    -2.17
    開心
    -2.16
     continúas
    -2.14
    POSITIVE LOGITS
     gantung
    3.06
    ࿙
    2.88
    玙
    2.81
    ſſel
    2.66
     bukanlah
    2.59
    他们在
    2.55
    你在
    2.52
     bendera
    2.50
     tetapi
    2.45
     теперь
    2.45
    Activations Density 0.002%

    No Known Activations