© 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. 34-GEMMASCOPE-RES-131K
    7. 92833
    Prev
    Next
    INDEX
    Explanations

    license or configuration details

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

    The neuron fires strongly on standalone numbers (especially multi‐digit tokens like years, version numbers, or other numeric codes).

    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
     Pregnancy
    -0.76
    شفى
    -0.75
    KAR
    -0.72
     Cartel
    -0.71
     Invention
    -0.70
     intercourse
    -0.68
     Sira
    -0.68
    メット
    -0.67
     TAT
    -0.66
     كريم
    -0.66
    POSITIVE LOGITS
    type
    0.75
     bisa
    0.73
     ofrecerte
    0.71
     TYPE
    0.68
     Type
    0.68
     Gaussian
    0.67
    Gaussian
    0.66
    TypeString
    0.65
    気持ち
    0.65
    чему
    0.65
    Activations Density 0.065%

    No Known Activations