© 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. 34-GEMMASCOPE-RES-131K
    7. 63858
    Prev
    Next
    INDEX
    Explanations

    titles or compound nouns

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

    This neuron detects license-related boilerplate terms, such as “All rights reserved” and “Redistribution.”

    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
    rania
    -0.78
     however
    -0.76
     юби
    -0.75
     isInitialized
    -0.75
     anyhow
    -0.75
    However
    -0.75
    それから
    -0.74
    ercicio
    -0.74
     interesante
    -0.73
     extraordinaria
    -0.72
    POSITIVE LOGITS
     kulak
    0.90
     Chilean
    0.90
    Portland
    0.89
    SEX
    0.88
     Clancy
    0.88
     Portland
    0.85
     BRIDGE
    0.83
     paralel
    0.82
    rors
    0.81
     Dij
    0.81
    Activations Density 0.001%

    No Known Activations