© 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. 22-GEMMASCOPE-RES-131K
    7. 15762
    Prev
    Next
    INDEX
    Explanations

    passionate fans and crowds

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

    The neuron is activated by passages describing crowd or fan presence and enthusiasm—mentions of attendance, stadium atmosphere, cheers, and spectators’ passion.

    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
     kroppen
    -1.08
    /#{
    -1.07
    available
    -1.05
     revered
    -1.05
    loit
    -1.03
    super
    -1.00
     huden
    -1.00
    self
    -1.00
     matéria
    -1.00
    wig
    -0.99
    POSITIVE LOGITS
     their
    1.29
    他们
    1.29
     support
    1.27
    :「
    1.21
     they
    1.19
    啬
    1.12
    他们在
    1.10
    他们的
    1.09
    ımda
    1.08
     támogat
    1.07
    Activations Density 0.026%

    No Known Activations