© 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. 10-GEMMASCOPE-RES-131K
    7. 130721
    Prev
    Next
    INDEX
    Explanations

    numbers like 37

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

    The neuron selectively responds to occurrences of the exact two‐digit number “37.”

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
     nové
    -2.44
    鼱
    -2.38
    zeppelin
    -2.36
    ならではの
    -2.36
    rolex
    -2.34
     informaron
    -2.33
    瘊
    -2.28
     mutfak
    -2.27
    瑆
    -2.23
     konkuren
    -2.22
    POSITIVE LOGITS
    ’
    3.30
    ↵
    3.30
    m
    3.11
    n
    3.02
    X
    2.88
    in
    2.86
    r
    2.77
    N
    2.72
    x
    2.70
    li
    2.59
    Activations Density 0.016%

    No Known Activations