© 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. 34-GEMMASCOPE-RES-131K
    7. 63724
    Prev
    Next
    INDEX
    Explanations

    pay-per-view events

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

    The neuron lights up on tokens that occur in pro‐wrestling/MMA event listings—organization acronyms (WWE, UFC), pay-per-view markers (PPV, “In Your House”), and the numbered editions or dates of those shows.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
    šel
    -0.86
    ăn
    -0.82
     周年
    -0.77
    istique
    -0.76
     procurar
    -0.75
     bukit
    -0.74
    ๔
    -0.72
    üne
    -0.71
     Monti
    -0.69
    Yellow
    -0.69
    POSITIVE LOGITS
     PPV
    1.60
    PP
    1.27
     event
    1.22
     pp
    1.16
    event
    1.05
     buys
    0.98
     PP
    0.97
     pay
    0.97
    BUY
    0.91
     ordering
    0.85
    Activations Density 0.028%

    No Known Activations