© 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. 16265
    Prev
    Next
    INDEX
    Explanations

    ESP32

    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
     for
    -1.51
     by
    -1.24
     when
    -1.19
     this
    -1.19
     to
    -1.13
     certainly
    -1.11
     understand
    -1.11
     should
    -1.09
     both
    -1.08
     after
    -1.02
    POSITIVE LOGITS
    Recomendaciones
    1.01
    馓
    1.00
    Beneficios
    1.00
     useAuth
    0.98
     discrepan
    0.98
     couvrir
    0.97
     dando
    0.96
     telles
    0.96
    踅
    0.95
    Screens
    0.94
    Activations Density 0.003%

    No Known Activations