© 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. 22-GEMMASCOPE-RES-131K
    7. 65974
    Prev
    Next
    INDEX
    Explanations

    whatever or call them what you will

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

    The neuron detects “all the … you want”–style promotional/listing phrases that offer an abundance of options.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
     Alen
    -0.97
    сього
    -0.96
    owana
    -0.94
     minta
    -0.94
    رافیک
    -0.93
     was
    -0.93
    tifier
    -0.92
     ͡°)
    -0.90
    }>
    -0.88
    вобо
    -0.84
    POSITIVE LOGITS
     but
    1.27
     whatever
    1.17
    whatever
    1.05
     Whatever
    1.05
     solange
    1.01
    Whatever
    0.95
    novu
    0.95
     gabinete
    0.94
    However
    0.93
     will
    0.93
    Activations Density 0.029%

    No Known Activations