© 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. 10-GEMMASCOPE-RES-131K
    7. 65380
    Prev
    Next
    INDEX
    Explanations

    names like paolo followed by names or locations, and 'catalyzes' or 'development'

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

    Italian first names, particularly "Paolo".

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi

    The neuron is triggered by occurrences of the name “Paolo.”

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
     That
    -2.48
    饧
    -2.44
    .
    -2.34
     добавля
    -2.25
     adicion
    -2.25
    によく
    -2.25
     และ
    -2.19
     possono
    -2.17
    ма
    -2.13
     acresc
    -2.13
    POSITIVE LOGITS
    In
    3.00
    我们
    2.75
    Unfortunately
    2.66
    Very
    2.45
    数十
    2.39
    With
    2.36
    Perhaps
    2.36
    There
    2.34
    When
    2.33
    Often
    2.33
    Activations Density 0.003%

    No Known Activations