© 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. 111191
    Prev
    Next
    INDEX
    Explanations

    early days/years

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

    This neuron detects references to an initial time period, especially the phrase “early days.”

    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
     gebouw
    -2.05
    顔を
    -1.91
     "
    -1.81
     maravillosa
    -1.80
     empêche
    -1.78
     détruit
    -1.76
    鉭
    -1.75
     cómoda
    -1.72
     diseñados
    -1.72
    一脸
    -1.70
    POSITIVE LOGITS
     of
    3.75
     This
    2.16
    w
    1.98
    f
    1.88
     incluso
    1.84
    j
    1.82
     их
    1.80
    手法
    1.73
    il
    1.70
    y
    1.70
    Activations Density 0.014%

    No Known Activations