© 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. 73439
    Prev
    Next
    INDEX
    Explanations

    code snippets and programming instructions

    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
     llevando
    -0.96
     viendo
    -0.93
     that
    -0.91
     tuvimos
    -0.91
     passou
    -0.88
     devem
    -0.88
     dispuesto
    -0.87
     Few
    -0.87
    𝗟
    -0.86
     recibiendo
    -0.85
    POSITIVE LOGITS
     lastly
    1.02
    kony
    0.98
    īgs
    0.95
     finally
    0.90
     scuole
    0.90
     compone
    0.89
     campion
    0.88
    küche
    0.88
    Infin
    0.86
     cuaca
    0.85
    Activations Density 0.099%

    No Known Activations