© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 27379
    Prev
    Next
    INDEX
    Explanations

    programming error handling and timing

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

    The neuron activates on exception-handling keywords (e.g. try, catch, except/recover) in code.

    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
     chacune
    -0.98
     AMEN
    -0.95
     Daya
    -0.91
     ランキング
    -0.90
     Magician
    -0.89
     officiels
    -0.89
     immediatamente
    -0.88
     fann
    -0.88
     konkre
    -0.88
    њи
    -0.88
    POSITIVE LOGITS
     time
    1.13
     ряда
    0.95
    izacja
    0.93
    помним
    0.92
     belieb
    0.87
    StandardCharsets
    0.86
     grà
    0.85
     Auckland
    0.84
    ских
    0.84
    ToMove
    0.84
    Activations Density 0.001%

    No Known Activations