© 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. 22-GEMMASCOPE-RES-131K
    7. 27635
    Prev
    Next
    INDEX
    Explanations

    SQL syntax errors

    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
     italiano
    -1.24
     for
    -1.15
     என்ப
    -1.14
     haut
    -1.11
    ارد
    -1.09
     bré
    -1.09
     alam
    -1.09
     tigre
    -1.05
     surat
    -1.04
     fraiche
    -1.04
    POSITIVE LOGITS
     “
    1.31
     "
    1.09
     "#
    1.05
     encender
    1.04
    楽しめます
    1.03
     preguntar
    1.02
     escoger
    1.02
     señalar
    1.02
     quintessential
    1.00
    究竟
    0.98
    Activations Density 0.009%

    No Known Activations