© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 74113
    Prev
    Next
    INDEX
    Explanations

    dogs and cats

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

    The neuron activates on words naming household pets—especially dogs (e.g. “dog,” “puppy,” “pooch,” 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
    ↵
    -1.92
     into
    -1.30
    le
    -1.29
    er
    -1.29
    wherein
    -1.27
    s
    -1.26
    eluaran
    -1.26
     świą
    -1.24
     ۱۱
    -1.21
    funcionario
    -1.19
    POSITIVE LOGITS
     our
    1.51
     two
    1.45
    ación
    1.30
     pegar
    1.27
     we
    1.26
     gravidanza
    1.26
    -------
    1.25
    ığımız
    1.24
     rehabilitación
    1.23
     две
    1.22
    Activations Density 0.026%

    No Known Activations