© 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. 128931
    Prev
    Next
    INDEX
    Explanations

    emergence of new schools or concepts

    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
     itali
    -1.13
    though
    -1.13
     REDUCTION
    -1.06
     truff
    -1.05
     existant
    -1.05
    every
    -1.05
     away
    -1.05
    鞫
    -1.04
     carbur
    -1.02
     its
    -1.01
    POSITIVE LOGITS
     new
    1.90
    新的
    1.48
     после
    1.14
     from
    1.13
     νέ
    1.11
     характеризу
    1.08
     neuen
    1.07
     after
    1.06
     makeshift
    1.06
     daugiau
    1.05
    Activations Density 0.079%

    No Known Activations