© 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. 10-GEMMASCOPE-RES-131K
    7. 69966
    Prev
    Next
    INDEX
    Explanations

    error conditions and markup

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

    The neuron fires on occurrences of the exception class name “InvalidArgumentException” (or the token “InvalidArgument” in exception identifiers).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
    of
    -2.56
    o
    -2.48
    ться
    -2.48
     â
    -2.41
     lament
    -2.41
     «
    -2.39
     -
    -2.36
    e
    -2.34
     OF
    -2.30
     pf
    -2.28
    POSITIVE LOGITS
    袮
    2.89
    ftrag
    2.56
    Ꮄ
    2.55
    鈇
    2.48
    利
    2.41
    雳
    2.41
    InvalidProtocol
    2.39
     Hinblick
    2.38
    )_
    2.31
    攙
    2.23
    Activations Density 0.000%

    No Known Activations