© 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. 34-GEMMASCOPE-RES-131K
    7. 30341
    Prev
    Next
    INDEX
    Explanations

    hypothetical examples or person names

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

    The neuron fires on words and symbols that introduce examples or hypothetical scenarios (e.g. “say,” “for example,” numeric values, “e.g.”, commas/colons in illustrative lists).

    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
    Mush
    -0.90
    yant
    -0.87
     glutathione
    -0.85
    任何
    -0.84
     Hough
    -0.84
     Insp
    -0.81
    日语
    -0.81
    ugía
    -0.80
     Polícia
    -0.79
    ❝
    -0.79
    POSITIVE LOGITS
     Johnny
    1.72
     person
    1.71
     Person
    1.66
     Joe
    1.62
     Jane
    1.52
     Sally
    1.50
     Bob
    1.49
     XYZ
    1.48
    Person
    1.48
     Mr
    1.47
    Activations Density 0.133%

    No Known Activations