© 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. 22-GEMMASCOPE-RES-131K
    7. 39779
    Prev
    Next
    INDEX
    Explanations

    qualified, promoted, entitled

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

    This neuron activates on terms describing sports‐competition progression, such as advancing, relegation, play‐offs, and related team outcomes.

    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
    嘻嘻
    -0.98
    stums
    -0.94
     "'";
    -0.92
    CardModule
    -0.91
    towane
    -0.86
    XMLSchema
    -0.84
     "'");
    -0.82
    czer
    -0.82
     constat
    -0.81
    specially
    -0.79
    POSITIVE LOGITS
     then
    1.24
     automatically
    1.05
     matic
    0.96
     eligible
    0.95
    まります
    0.92
     automatiquement
    0.89
     entitled
    0.88
     автоматически
    0.87
     Gesamt
    0.87
    brite
    0.87
    Activations Density 0.065%

    No Known Activations