© 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. 118301
    Prev
    Next
    INDEX
    Explanations

    titles for women

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

    The neuron primarily fires on honorifics (“Mr,” “Mrs,” etc.) and the coordinating conjunction “and” as they appear in formal listings of people.

    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
     ө
    -0.81
    itarianism
    -0.79
     riêng
    -0.76
    排除
    -0.74
     legales
    -0.74
    couz
    -0.73
     regionales
    -0.73
    entennial
    -0.71
     esenciales
    -0.71
    ghold
    -0.71
    POSITIVE LOGITS
     Mrs
    1.26
    Mrs
    1.07
     mrs
    0.95
     zustand
    0.89
     pomeriggio
    0.82
     signora
    0.81
     doña
    0.81
     Misses
    0.80
    اں
    0.79
     Pty
    0.77
    Activations Density 0.013%

    No Known Activations