© 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. 34-GEMMASCOPE-RES-131K
    7. 94171
    Prev
    Next
    INDEX
    Explanations

    questions about how when or what

    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_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
     nếu
    -0.94
     após
    -0.88
     enquanto
    -0.84
     урока
    -0.80
     dn
    -0.77
     يمكنك
    -0.76
     ihren
    -0.75
    在你
    -0.75
     gdyby
    -0.73
    onnen
    -0.72
    POSITIVE LOGITS
     why
    2.03
     Who
    1.98
     How
    1.94
     Why
    1.92
     HOW
    1.91
    how
    1.87
     Where
    1.83
    Where
    1.80
     how
    1.80
    Who
    1.77
    Activations Density 0.051%

    No Known Activations