© 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. 34-GEMMASCOPE-RES-131K
    7. 99081
    Prev
    Next
    INDEX
    Explanations

    errors and exceptions

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

    The main thing this neuron does is detect mentions of errors or exceptions (e.g., ValidationError, errors, exception, report).

    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
     vicino
    -0.81
     solchen
    -0.77
    必要
    -0.77
     Jé
    -0.74
     conditions
    -0.73
     国際
    -0.72
     entsprechend
    -0.72
    Special
    -0.71
     Iber
    -0.71
    いますが
    -0.70
    POSITIVE LOGITS
     any
    0.84
     erase
    0.80
    cocci
    0.79
    Affirmed
    0.79
    leaved
    0.79
    Aggregator
    0.79
    kker
    0.76
    Dessert
    0.76
    tro
    0.75
    ját
    0.75
    Activations Density 0.007%

    No Known Activations