© 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. 34-GEMMASCOPE-RES-131K
    7. 59576
    Prev
    Next
    INDEX
    Explanations

    makeup artist, bridal, hair

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

    The neuron fires strongly on tokens ending in the suffix “-ist,” i.e. occupation or role names such as harpist, pianist, artist, 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_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
    propagate
    -0.75
    Developer
    -0.70
     inneces
    -0.70
    eaza
    -0.69
     وصف
    -0.68
     científico
    -0.68
     knitted
    -0.68
    arah
    -0.67
     Wier
    -0.66
     トート
    -0.66
    POSITIVE LOGITS
     makeup
    2.30
     Makeup
    1.85
    makeup
    1.84
    Makeup
    1.71
     hair
    1.52
     bridal
    1.30
     Hair
    1.25
     hairst
    1.22
    メイク
    1.21
    mua
    1.19
    Activations Density 0.009%

    No Known Activations