© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 28299
    Prev
    Next
    INDEX
    Explanations

    numbers followed by commas

    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
     what
    -2.25
     which
    -2.19
     that
    -1.93
     this
    -1.84
     unique
    -1.68
     truly
    -1.53
     both
    -1.52
     really
    -1.45
     who
    -1.44
     mostly
    -1.44
    POSITIVE LOGITS
     solides
    1.51
    idän
    1.44
     vorige
    1.39
     Strukt
    1.39
    Ive
    1.38
     hef
    1.36
    lil
    1.33
     odes
    1.30
    halle
    1.30
     tirs
    1.30
    Activations Density 0.014%

    No Known Activations