© 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. 34-GEMMASCOPE-RES-131K
    7. 119327
    Prev
    Next
    INDEX
    Explanations

    decades and years

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

    The neuron is highly sensitive to numeric characters or sequences (individual digits and multi‐digit numbers) in the text.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
     ۲
    -0.75
     raum
    -0.73
     Patterson
    -0.72
     opção
    -0.72
    utando
    -0.69
    arme
    -0.69
    repositories
    -0.69
     heritage
    -0.69
    مانية
    -0.69
     konkurs
    -0.68
    POSITIVE LOGITS
     década
    1.23
     decade
    1.10
    decade
    1.06
     Decade
    0.87
    anggan
    0.80
    prefixes
    0.78
    interactions
    0.78
    сё
    0.77
    Prefix
    0.77
    acteria
    0.77
    Activations Density 0.058%

    No Known Activations