© 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. 128257
    Prev
    Next
    INDEX
    Explanations

    requirements and demands

    np_acts-logits-general · gemini-2.5-flash-lite
    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
     and
    -0.88
     destinados
    -0.81
     Njema
    -0.80
    -0.79
    prende
    -0.78
    ñada
    -0.77
     начал
    -0.77
    intended
    -0.76
    angle
    -0.75
    BufferedWriter
    -0.75
    POSITIVE LOGITS
     requirements
    1.56
     requirement
    1.52
     requiring
    1.44
     require
    1.41
    要求
    1.28
     permintaan
    1.23
     demand
    1.23
     requieren
    1.20
     requ
    1.18
     needs
    1.08
    Activations Density 0.081%

    No Known Activations