© 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. 15714
    Prev
    Next
    INDEX
    Explanations

    vertical and horizontal layout

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

    The neuron fires on code‐style constructor or enum/class instantiation tokens (e.g. “new”, class names, enum references) rather than normal words.

    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
    geschenk
    -1.10
    🚼
    -1.00
    がなく
    -0.99
    そして
    -0.98
    瘊
    -0.98
     algumas
    -0.98
     keten
    -0.96
    与此同时
    -0.95
    ouchable
    -0.95
    ಃ
    -0.94
    POSITIVE LOGITS
     so
    1.14
     will
    1.05
     strangely
    1.04
     They
    1.01
     вони
    1.00
     düny
    1.00
     may
    0.98
     aptly
    0.97
     Before
    0.97
     very
    0.97
    Activations Density 0.001%

    No Known Activations