© 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. 26105
    Prev
    Next
    INDEX
    Explanations

    contrasts and conjunctions

    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
     három
    -1.53
     megfelelő
    -1.47
     sogenannten
    -1.44
     Если
    -1.41
     denominado
    -1.37
     største
    -1.35
     azonban
    -1.35
     grecque
    -1.30
     tzw
    -1.29
     henvis
    -1.28
    POSITIVE LOGITS
     it
    1.63
     and
    1.41
     to
    1.29
     others
    1.27
     for
    1.23
     especially
    1.22
    方針
    1.19
    因为它
    1.17
     then
    1.16
    要因
    1.16
    Activations Density 0.020%

    No Known Activations