© 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
    2. Gemma-2-27B
    3. 22-GEMMASCOPE-RES-131K
    4. 15978
    Prev
    Next
    INDEX
    Explanations

    previous values

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

    The neuron is picking out standalone numerical values (e.g. digits, decimals, percentages) 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_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
    ють
    -0.93
    ülő
    -0.88
     paillettes
    -0.85
     eier
    -0.85
     hunne
    -0.85
    NOAA
    -0.84
     eenen
    -0.84
    브
    -0.81
    第
    -0.81
    unless
    -0.81
    POSITIVE LOGITS
     previous
    1.41
     предыду
    1.15
     prior
    1.11
     précédente
    1.10
     previously
    1.04
     sebelumnya
    1.02
    時には
    1.00
    previous
    0.98
    もありました
    0.96
    も同じ
    0.95
    Activations Density 0.003%

    No Known Activations