© 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. 44339
    Prev
    Next
    INDEX
    Explanations

    watermark mark border stain

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

    The neuron fires on occurrences of “watermark” (and its variants like “watermarking”) in text.

    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
    2
    -2.13
    但是
    -1.99
     But
    -1.99
     avevo
    -1.94
    6
    -1.91
    高达
    -1.89
     Also
    -1.86
    4
    -1.85
    AtIndexPath
    -1.83
    al
    -1.82
    POSITIVE LOGITS
    小结
    2.44
    𓆜
    2.25
    伱
    2.05
    趼
    2.02
    臜
    2.02
    此时的
    2.00
    </strong>
    2.00
    
    1.95
    实战
    1.92
    産の
    1.91
    Activations Density 0.012%

    No Known Activations