© 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. 10-GEMMASCOPE-RES-131K
    7. 112860
    Prev
    Next
    INDEX
    Explanations

    UNITED STATES

    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_10/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
     keeps
    -1.84
     before
    -1.70
     showed
    -1.62
     one
    -1.61
     and
    -1.59
     although
    -1.56
     kept
    -1.54
     this
    -1.52
     allows
    -1.52
     wants
    -1.47
    POSITIVE LOGITS
     VERY
    1.72
    藐
    1.47
    —–
    1.44
    hopped
    1.41
     delightfully
    1.41
     quintessential
    1.41
    ¾
    1.41
     grumpy
    1.41
     Franse
    1.40
    provided
    1.38
    Activations Density 0.092%

    No Known Activations