© 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. 34-GEMMASCOPE-RES-131K
    7. 38650
    Prev
    Next
    INDEX
    Explanations

    names of people

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

    The neuron is especially triggered by capitalized name tokens and initials (e.g. single‐letter initials and proper names) in academic or citation contexts.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
    ARROLL
    -0.81
     Creative
    -0.74
    ʐ
    -0.71
    ^(@
    -0.70
    configura
    -0.70
    StringTo
    -0.69
    )%
    -0.68
    еня
    -0.66
    fieur
    -0.65
     sungai
    -0.65
    POSITIVE LOGITS
    Comple
    0.85
    Ain
    0.77
     completar
    0.76
    Reflect
    0.73
     حقوق
    0.72
    Ao
    0.68
    rophobic
    0.68
    Ax
    0.68
     где
    0.68
    Mayor
    0.68
    Activations Density 0.019%

    No Known Activations