© 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
    2. Gemma-2-27B
    3. 34-GEMMASCOPE-RES-131K
    4. 60404
    Prev
    Next
    INDEX
    Explanations

    numbers followed by words or dates

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

    The neuron strongly activates on numeric tokens—especially years, dates, and edition 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_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
     Param
    -0.85
     Interpre
    -0.78
     汤
    -0.77
    amiento
    -0.75
    ת
    -0.74
     europeo
    -0.73
     Nicaragua
    -0.72
     Sherwood
    -0.72
    セット
    -0.71
     Gilbert
    -0.71
    POSITIVE LOGITS
     ortak
    0.86
     речь
    0.78
    expectations
    0.75
     co
    0.74
    できて
    0.74
    Expectations
    0.69
     mortos
    0.68
    barber
    0.68
    pozycji
    0.66
     Expectations
    0.66
    Activations Density 0.023%

    No Known Activations