© 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. 22-GEMMASCOPE-RES-131K
    7. 65744
    Prev
    Next
    INDEX
    Explanations

    favorite books and movies

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

    The neuron is specialized in detecting numeric and quantitative references—numbers, counts, statistics, and similar measures 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
    6
    -1.05
    is
    -0.98
    儀式
    -0.96
     is
    -0.95
    了一眼
    -0.92
     instig
    -0.90
    5
    -0.90
    りが
    -0.90
    ani
    -0.88
    ayage
    -0.88
    POSITIVE LOGITS
     popular
    1.10
     Lieblings
    1.09
     berjudul
    1.07
     movie
    1.05
     selected
    1.05
     choisi
    1.04
    《
    1.04
    favorite
    1.04
     Aufstieg
    1.04
     Dě
    1.03
    Activations Density 0.138%

    No Known Activations