© 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. 7298
    Prev
    Next
    INDEX
    Explanations

    kindness and compassion

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

    The main thing this neuron does is detect words expressing kindness, compassion, or caring (e.g. “kind,” “compassionate,” “kinder,” etc.).

    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
    GETHER
    -1.57
    нения
    -1.50
    нение
    -1.47
    ৄ
    -1.47
    ि
    -1.45
    abbix
    -1.38
    нии
    -1.38
    ributor
    -1.38
     который
    -1.36
    ляем
    -1.35
    POSITIVE LOGITS
    7
    2.16
    9
    2.06
    5
    1.95
    6
    1.77
    Those
    1.77
    8
    1.77
    4
    1.62
     animali
    1.58
    These
    1.49
    她的
    1.45
    Activations Density 0.046%

    No Known Activations