© 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. 10-GEMMASCOPE-RES-131K
    7. 113405
    Prev
    Next
    INDEX
    Explanations

    code comments with explanations

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

    the opening brace of a code block or function body in programming code.

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

    The neuron is detecting leading whitespace indentation (tabs or spaces) at the beginnings of code lines.

    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
     for
    -2.25
     all
    -2.05
     Despite
    -2.03
     revital
    -1.95
     an
    -1.95
     gives
    -1.93
     also
    -1.91
     burgeoning
    -1.88
     provide
    -1.88
     invigor
    -1.88
    POSITIVE LOGITS
     VERY
    2.02
     astounding
    1.92
     puramente
    1.91
     startling
    1.90
     !!!!
    1.88
    擼
    1.87
     lauded
    1.84
     unbelievably
    1.84
     différents
    1.83
     usando
    1.83
    Activations Density 0.156%

    No Known Activations