© 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. 48727
    Prev
    Next
    INDEX
    Explanations

    partnership, singles, runs, momentum

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

    This neuron detects numeric statistics and scores in cricket match contexts.

    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
     menyeb
    -0.84
     Liban
    -0.83
     Levine
    -0.83
    тическое
    -0.82
    ugen
    -0.82
    강
    -0.81
    reported
    -0.81
     朱
    -0.80
     ()
    -0.79
    ílica
    -0.79
    POSITIVE LOGITS
     partnership
    1.05
     Partnership
    0.99
     singles
    0.92
     partnerships
    0.91
     pressure
    0.89
    Partnership
    0.89
     Partnerships
    0.85
     restrict
    0.84
     restrictions
    0.83
     momentum
    0.82
    Activations Density 0.006%

    No Known Activations