© 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. 57636
    Prev
    Next
    INDEX
    Explanations

    abstract noun suffixes

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

    This neuron fires on long, multi-syllable stems—especially Greek/Latin–derived roots or other formal technical word beginnings.

    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
    roso
    -1.02
     investigating
    -0.97
     możli
    -0.93
     войска
    -0.92
     negotiating
    -0.91
    knights
    -0.91
    шаем
    -0.90
     ‘‘
    -0.89
     offerings
    -0.88
    parfum
    -0.88
    POSITIVE LOGITS
     of
    1.81
    ization
    1.68
    isation
    1.57
    ishment
    1.38
    ulation
    1.33
    ification
    1.32
    fication
    1.30
    ylation
    1.25
    ation
    1.20
    duction
    1.19
    Activations Density 0.124%

    No Known Activations