© 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. 68981
    Prev
    Next
    INDEX
    Explanations

    technical code and french nouns

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

    The neuron fires on numeric literals (numbers) in code or markup.

    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
     With
    -1.13
     also
    -1.13
     their
    -1.05
     even
    -0.98
     do
    -0.96
     составля
    -0.94
     why
    -0.94
     with
    -0.93
     If
    -0.92
     help
    -0.88
    POSITIVE LOGITS
     composés
    1.28
     célè
    1.27
     kommune
    1.25
     chaus
    1.21
     genoux
    1.21
     peluche
    1.21
     médic
    1.21
     menant
    1.20
     rése
    1.20
     flott
    1.20
    Activations Density 0.158%

    No Known Activations