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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 41634
    Prev
    Next
    INDEX
    Explanations

    Late in time or season

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

    This neuron activates primarily on occurrences of the word “late” (often capitalized) as a temporal marker.

    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
      
    -3.39
    ’
    -3.28
    ↵↵
    -3.05
     zwölf
    -2.69
    .’”
    -2.52
    迨
    -2.48
     nachfol
    -2.47
    <
    -2.36
     t
    -2.33
     of
    -2.27
    POSITIVE LOGITS
    也是
    2.69
    Setelah
    2.69
    Даже
    2.63
    Конечно
    2.56
    Selain
    2.45
    Хотя
    2.44
     マンション
    2.41
    Bahkan
    2.41
     inoxydable
    2.39
    e
    2.39
    Activations Density 0.009%

    No Known Activations