© 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. 103360
    Prev
    Next
    INDEX
    Explanations

    foreign country names and places

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

    This neuron responds strongly to isolated bits of reference‐style metadata—short letter or number tokens, domain names, initials, and other small alphanumeric pieces you’d see in citations, URLs, and headings.

    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
     gani
    -1.13
     kuu
    -1.05
    درمان
    -0.99
    desde
    -0.98
     wszystkich
    -0.92
    چگونه
    -0.92
     zarówno
    -0.91
     wszystkie
    -0.90
    rahat
    -0.90
    SpringBoot
    -0.89
    POSITIVE LOGITS
     находя
    1.02
     weaknesses
    0.94
     Bournemouth
    0.93
     βρί
    0.93
    hicles
    0.92
    posicao
    0.92
     breaths
    0.91
    bilidad
    0.90
     congratulations
    0.90
     Statt
    0.89
    Activations Density 0.000%

    No Known Activations