© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 65701
    Prev
    Next
    INDEX
    Explanations

    how to name things

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

    This neuron detects contexts where the text is asking for or referring to “names” or the act of naming (e.g. naming a range, column, class, file, or window).

    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
     all
    -1.27
    inerja
    -1.20
    Other
    -1.20
     extenso
    -1.08
     autre
    -1.08
     tradicional
    -1.07
     and
    -1.03
    所有
    -1.00
    {
    -0.99
    各
    -0.98
    POSITIVE LOGITS
     then
    1.66
     kemudian
    1.48
     muitos
    1.43
     ciencias
    1.39
     Then
    1.31
     continúa
    1.30
    ELTS
    1.27
     muitas
    1.25
     danach
    1.25
     llamó
    1.24
    Activations Density 0.110%

    No Known Activations