© 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. 34-GEMMASCOPE-RES-131K
    7. 33210
    Prev
    Next
    INDEX
    Explanations

    best practices

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

    The neuron flags occurrences of the word “practice” (especially in the phrase “best practices”).

    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
     best
    -1.44
    best
    -1.30
     najlepiej
    -0.96
     BEST
    -0.95
    ree
    -0.85
    BEST
    -0.82
    Best
    -0.80
     refresher
    -0.76
    評價
    -0.76
     Penal
    -0.75
    POSITIVE LOGITS
     practices
    3.52
     Practices
    2.64
    practices
    2.50
     practice
    2.45
    Practices
    2.42
     PRACTICES
    2.13
    practice
    1.95
     Practice
    1.91
     pratiques
    1.87
    Practice
    1.78
    Activations Density 0.029%

    No Known Activations