© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 103128
    Prev
    Next
    INDEX
    Explanations

    numbers as measurements

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

    The neuron fires on numeric tokens, i.e. standalone digits or multi-digit numbers (measurements, model numbers, percentages, etc.).

    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
     
    -2.44
    ↵
    -1.93
     new
    -1.74
     during
    -1.67
     after
    -1.66
    :
    -1.41
     change
    -1.38
     (
    -1.38
    它
    -1.35
     после
    -1.34
    POSITIVE LOGITS
     apartament
    1.73
     interieur
    1.66
     guapos
    1.52
    ţe
    1.51
    sonaro
    1.47
     pantal
    1.44
     bont
    1.41
    猡
    1.41
     alltid
    1.39
     calentamiento
    1.39
    Activations Density 0.074%

    No Known Activations