© 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
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 11449
    Prev
    Next
    INDEX
    Explanations

    boolean values and semicolons

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

    the boolean literal "true" in programming code.

    oai_token-act-pair · claude-4-5-sonnetTriggered by @bfgcn7frf5

    The neuron detects code patterns where a variable is being assigned a Boolean literal (e.g. “x = true” or “x = false”).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
     menyukai
    -1.49
     menyadari
    -1.39
     heureuse
    -1.37
     setuju
    -1.34
     merasakan
    -1.31
     remporté
    -1.27
     váš
    -1.27
     artikkeli
    -1.26
     vaše
    -1.25
     surpren
    -1.22
    POSITIVE LOGITS
     instead
    1.88
     its
    1.69
    );
    1.44
     then
    1.41
    </h2>
    1.37
    TestingModule
    1.34
     couldn
    1.33
    <h2>
    1.32
    >
    1.30
     Instead
    1.30
    Activations Density 0.008%

    No Known Activations