© 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. 36150
    Prev
    Next
    INDEX
    Explanations

    3D space coordinates

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

    The neuron detects numeric tokens (numerals) in the text, with stronger activation for larger multi‐digit 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
    план
    -0.98
     yetu
    -0.94
     trasferimento
    -0.91
    refront
    -0.90
    OREM
    -0.90
    Vere
    -0.88
     比較
    -0.85
    ':{'
    -0.85
     satisfacción
    -0.84
     professionals
    -0.84
    POSITIVE LOGITS
     voxel
    1.08
    ffey
    0.91
     unwillingness
    0.90
    scul
    0.90
    space
    0.88
    Prism
    0.88
    entist
    0.87
    gettes
    0.85
    zember
    0.85
    ."/
    0.85
    Activations Density 0.138%

    No Known Activations