© 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. 82437
    Prev
    Next
    INDEX
    Explanations

    sports leagues and tournaments

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

    The neuron is sensitive to numeric tokens, especially multi‐digit numbers such as years, scores, statistics, or other numeric values.

    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
    anggaran
    -1.05
    enged
    -0.94
     endast
    -0.84
     ihan
    -0.83
    ḷ
    -0.82
     Ciências
    -0.81
     kuliner
    -0.81
    ượng
    -0.81
    astă
    -0.81
     Pakistani
    -0.80
    POSITIVE LOGITS
     Vương
    0.85
    absolute
    0.84
    oia
    0.84
    ůž
    0.83
     Direc
    0.82
    ぼ
    0.81
     Thiên
    0.81
    扯
    0.81
    looking
    0.81
    bell
    0.79
    Activations Density 0.005%

    No Known Activations