© 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. 91376
    Prev
    Next
    INDEX
    Explanations

    one-on-one and personal

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

    The neuron fires on phrases describing one-on-one or highly personalized interactions (e.g., “one-on-one,” “working with clients,” “collaborating with you”).

    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
    lai
    -0.79
     Pente
    -0.77
    ろう
    -0.75
     Late
    -0.74
    inn
    -0.73
    equ
    -0.73
     Veil
    -0.73
    様が
    -0.71
     ngay
    -0.71
    素直
    -0.71
    POSITIVE LOGITS
     individualized
    2.22
     индивидуаль
    2.02
     individually
    2.00
     individual
    1.95
     one
    1.87
    Individual
    1.80
    individual
    1.79
     indywidual
    1.79
     индивиду
    1.78
     personalized
    1.77
    Activations Density 0.029%

    No Known Activations