© 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. 10-GEMMASCOPE-RES-131K
    7. 39633
    Prev
    Next
    INDEX
    Explanations

    holder

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

    The neuron activates on occurrences of the word “holder” (often as a suffix in compounds like card-holder, ticket-holder, mortgage-holder, 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_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
    x
    -3.14
    v
    -3.05
    y
    -2.73
    であった
    -2.45
    и
    -2.36
    vantaged
    -2.34
    ’
    -2.33
    e
    -2.30
     '
    -2.25
    }$
    -2.23
    POSITIVE LOGITS
    淠
    3.30
     аккура
    2.94
    焄
    2.78
    ⟞
    2.67
    鱂
    2.50
    ࿊
    2.50
    
    2.48
    䊏
    2.45
    
    2.45
    犼
    2.44
    Activations Density 0.004%

    No Known Activations