© 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. 54341
    Prev
    Next
    INDEX
    Explanations

    words ending in -lized or -ed

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

    The neuron activates on numeric tokens (decimal or integer measurements) in the text.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    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
     comienzos
    -1.09
    厲
    -1.01
    nvm
    -0.99
     creó
    -0.94
     habitaciones
    -0.94
     inside
    -0.94
     visualización
    -0.93
     dormitorios
    -0.92
     fermeture
    -0.92
     filós
    -0.91
    POSITIVE LOGITS
    ETHING
    1.06
     continua
    0.93
     leksikon
    0.92
     انواع
    0.88
    ︠
    0.87
     obliga
    0.86
    共和国
    0.86
     just
    0.84
    opheles
    0.84
     it
    0.83
    Activations Density 0.159%

    No Known Activations