© 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. 131024
    Prev
    Next
    INDEX
    Explanations

    typical followed by examples

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

    The neuron fires on occurrences of the word “Typical” or its lowercase form “typically,” i.e., when text introduces standard or customary examples.

    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
     "
    -3.67
     The
    -3.19
     '
    -3.13
    ↵↵
    -3.00
     fünf
    -2.88
    了出来
    -2.81
     dando
    -2.78
     is
    -2.78
                
    -2.61
     ""
    -2.45
    POSITIVE LOGITS
    塤
    2.98
     diligent
    2.78
    месте
    2.63
    Even
    2.61
     rigorous
    2.59
     increases
    2.59
     meticulous
    2.56
     intensely
    2.53
     أيض
    2.53
    at
    2.53
    Activations Density 0.008%

    No Known Activations