© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 65926
    Prev
    Next
    INDEX
    Explanations

    never followed by various words

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

    This neuron responds to the appearance of the word “never.”

    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
    =
    -2.13
    ܞ
    -2.13
    茈
    -2.13
     artisti
    -2.11
    衕
    -2.09
     sosi
    -2.05
    ܜ
    -2.03
    }$.
    -2.03
     faſt
    -2.00
    俛
    -1.98
    POSITIVE LOGITS
     проєкту
    1.88
     
    1.87
     Не
    1.85
    ':
    1.82
     zarówno
    1.78
    制的
    1.77
    </i>
    1.75
     aquellos
    1.73
     jamás
    1.73
     Absolutely
    1.72
    Activations Density 0.032%

    No Known Activations