© 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. 34-GEMMASCOPE-RES-131K
    7. 106954
    Prev
    Next
    INDEX
    Explanations

    import "errors"

    np_acts-logits-general · gemini-2.5-flash-lite

    This neuron fires on numeric tokens (and their adjacent punctuation), e.g. digits in version numbers, chapter : verse citations, and standalone semicolons/colons.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
     levy
    -0.78
    getSupport
    -0.73
     script
    -0.69
     Beschreibung
    -0.67
     Nordrhein
    -0.66
     Sucher
    -0.65
    Princeton
    -0.64
    aura
    -0.64
     Script
    -0.63
     Prince
    -0.63
    POSITIVE LOGITS
     repositories
    0.84
     refus
    0.81
     bertahan
    0.69
     wren
    0.68
     remplissage
    0.66
    յ
    0.66
    enzio
    0.66
     polygon
    0.65
     Wilde
    0.65
     marquis
    0.65
    Activations Density 0.121%

    No Known Activations