© 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. 34-GEMMASCOPE-RES-131K
    7. 117349
    Prev
    Next
    INDEX
    Explanations

    error handling and exceptions

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

    the neuron detects error‐ and exception‐related tokens (e.g. “error,” “Exception,” “catch,” “uncaught,” etc.).

    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
     cảm
    -0.88
     acesta
    -0.88
    ™,
    -0.82
    reya
    -0.76
     phenomena
    -0.75
     viable
    -0.74
     involved
    -0.73
     rarely
    -0.72
    European
    -0.72
     Badminton
    -0.72
    POSITIVE LOGITS
    expected
    1.09
    otec
    0.88
    correct
    0.87
     correct
    0.85
     cualquier
    0.84
    WithFormat
    0.84
    正确的
    0.84
     správ
    0.83
     Expected
    0.82
    exactly
    0.81
    Activations Density 0.006%

    No Known Activations