© 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. 10370
    Prev
    Next
    INDEX
    Explanations

    verb "to do"

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

    The neuron strongly activates on the verb “do” (and its inflected forms like “doing” or “did”).

    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
     Municipalidad
    -1.59
     you
    -1.59
    ⅔
    -1.58
    ↵
    -1.55
     they
    -1.55
     is
    -1.51
    ⅜
    -1.48
    mannschaft
    -1.48
     simplemente
    -1.45
    ⅛
    -1.41
    POSITIVE LOGITS
    Contexte
    1.44
     porcelaine
    1.42
    Taille
    1.34
    lceil
    1.33
    jelasan
    1.33
    níky
    1.31
    Ici
    1.24
    嗐
    1.24
    ссии
    1.24
     an
    1.24
    Activations Density 0.029%

    No Known Activations