© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 38497
    Prev
    Next
    INDEX
    Explanations

    API endpoints and URLs

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

    The neuron activates on tokens containing numerals (i.e. digits), picking up numeric literals or identifiers with embedded 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
    caped
    -0.98
    poon
    -0.94
     Gestión
    -0.90
    却是
    -0.90
    λικά
    -0.89
    syst
    -0.89
    沙发
    -0.89
     him
    -0.88
    crates
    -0.88
     üzer
    -0.87
    POSITIVE LOGITS
     you
    1.05
     your
    0.98
     < 
    0.97
    𝒙
    0.96
     inzwischen
    0.94
     enfe
    0.93
    を中心
    0.92
     cinematogra
    0.91
    栩
    0.90
    álisis
    0.90
    Activations Density 0.006%

    No Known Activations