© 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. 77883
    Prev
    Next
    INDEX
    Explanations

    sports figures and events

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

    The neuron strongly activates on numeric tokens—scores, statistics, years, and other numbers—in sports‐style reporting.

    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
     penggemar
    -0.98
     educators
    -0.90
    老师
    -0.87
     penonton
    -0.84
    亹
    -0.82
     ซึ่ง
    -0.82
    anguages
    -0.82
    employer
    -0.81
    ínez
    -0.81
    ENNA
    -0.80
    POSITIVE LOGITS
     midfielder
    2.81
     winger
    2.42
     striker
    2.36
     forward
    2.33
     goalkeeper
    2.09
     center
    2.09
     guitarist
    2.06
     defender
    1.98
     bassist
    1.90
     centre
    1.87
    Activations Density 0.157%

    No Known Activations