© 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. 22-GEMMASCOPE-RES-131K
    4. 75329
    Prev
    Next
    INDEX
    Explanations

    exception type checking

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

    The neuron detects exception‐handling “catch” clauses—that is, the “catch” keyword, its opening parenthesis, and the exception type inside.

    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
     årene
    -1.16
     sini
    -1.14
     nyeri
    -1.08
    klere
    -1.06
     slike
    -1.06
    jsonObj
    -1.05
     GRATU
    -1.03
     kväll
    -1.02
     menant
    -1.02
    국의
    -1.02
    POSITIVE LOGITS
     anything
    1.61
     any
    1.41
     either
    1.32
     or
    1.23
    farlane
    1.14
     everything
    1.06
    any
    1.05
     glaubte
    1.01
     }),
    1.01
     things
    0.98
    Activations Density 0.001%

    No Known Activations