© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 26210
    Prev
    Next
    INDEX
    Explanations

    quantifiable measurements and statistics

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

    The neuron specializes in spotting numeric expressions—digits, numeral words, and related number-of-something tokens.

    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
     cristian
    -0.94
    стере
    -0.94
     regia
    -0.88
    пози
    -0.80
    ★★★★☆
    -0.79
     $$$
    -0.78
    합니다
    -0.77
     biografia
    -0.77
     contempl
    -0.75
     repeti
    -0.75
    POSITIVE LOGITS
    thodoxy
    1.02
    taya
    1.00
     takiej
    0.97
    zit
    0.95
     such
    0.93
    "|
    0.92
    Adapun
    0.90
    を除
    0.89
    erunner
    0.88
    *);
    0.88
    Activations Density 0.119%

    No Known Activations