© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 73845
    Prev
    Next
    INDEX
    Explanations

    exception or error inheritance

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

    The neuron responds to exception‐class definitions or usages—particularly to the token “Exception” (and related inheritance syntax like “< Exception”).

    oai_token-act-pair · o4-mini
    Triggered 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
    Negative Logits
    😊
    -0.96
     valentin
    -0.93
     насеља
    -0.93
    又是一
    -0.90
     科技
    -0.88
    ⇝
    -0.88
    paksa
    -0.87
    鲠
    -0.87
     blumen
    -0.86
     Rá
    -0.85
    POSITIVE LOGITS
    (!__
    1.05
    させていただきました
    1.00
     inability
    0.98
     זיי
    0.97
    //</
    0.95
     cases
    0.94
    太大
    0.93
     amikor
    0.93
    anything
    0.91
    oznam
    0.90
    Activations Density 0.019%

    No Known Activations