© 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. 22-GEMMASCOPE-RES-131K
    7. 34493
    Prev
    Next
    INDEX
    Explanations

    career and time periods

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

    The neuron fires on numeric and ranking expressions—cardinal or ordinal numbers, stats, and words indicating positions or metrics (e.g. “up,” “first,” “best,” “two,” “fourth”).

    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
     deux
    -1.00
    Things
    -0.91
    umina
    -0.91
    οντας
    -0.91
     måned
    -0.90
    えると
    -0.90
     Añade
    -0.89
     freio
    -0.89
     Hacer
    -0.88
    things
    -0.88
    POSITIVE LOGITS
     years
    1.41
     decades
    1.24
     tonight
    1.23
     this
    1.19
     lifetime
    1.09
     today
    1.09
     adulthood
    1.05
     careers
    1.02
     decade
    0.95
     lifelong
    0.94
    Activations Density 0.093%

    No Known Activations