© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 13894
    Prev
    Next
    INDEX
    Explanations

    things that seem untrue

    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
    ший
    -0.90
     terbukti
    -0.81
     необходимо
    -0.79
    itsa
    -0.78
     need
    -0.77
     Knoblauch
    -0.77
    مرار
    -0.76
    logeny
    -0.76
    щик
    -0.76
    喊道
    -0.75
    POSITIVE LOGITS
     disbelief
    1.23
     improbable
    1.11
     unlikely
    1.09
     inconceivable
    1.02
     таком
    0.98
     contradicts
    0.98
    unlikely
    0.92
    まさか
    0.90
     дописа
    0.89
     Doub
    0.88
    Activations Density 0.056%

    No Known Activations