© 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. 65975
    Prev
    Next
    INDEX
    Explanations

    import and using statements

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

    The neuron fires on standalone numeric literal tokens (numbers).

    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
    が
    -1.07
     summari
    -1.05
     tranf
    -1.00
     vorsichtig
    -0.99
    难怪
    -0.98
     repul
    -0.96
    说实话
    -0.96
    rindo
    -0.95
    érios
    -0.95
     anklicken
    -0.93
    POSITIVE LOGITS
     opérés
    1.05
     also
    1.01
     anſ
    0.99
    dienne
    0.98
    beit
    0.97
    cedent
    0.97
     privés
    0.96
     beslut
    0.90
     pintada
    0.90
     exklu
    0.90
    Activations Density 0.021%

    No Known Activations