© 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. 15821
    Prev
    Next
    INDEX
    Explanations

    additional benefits

    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
    chromatic
    -0.98
     there
    -0.94
     recently
    -0.93
    cobacterium
    -0.91
    lask
    -0.89
    ˅
    -0.88
    是这样的
    -0.87
    taxia
    -0.87
     kommunizieren
    -0.86
    ૮
    -0.86
    POSITIVE LOGITS
     dessutom
    1.20
     plus
    1.07
    न्द
    1.06
     navíc
    1.05
    了口气
    1.04
     får
    1.01
     gainera
    1.01
    /*",
    1.00
     וגם
    0.99
     benefit
    0.98
    Activations Density 0.037%

    No Known Activations