© 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-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 4989
    Prev
    Next
    INDEX
    Explanations

    error handling and conditions

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

    The neuron strongly activates on numeric tokens containing a decimal point—that is, floating-point literals.

    oai_token-act-pair · o4-mini
    Triggered by @jyhe0408

    lines of code or documentation where an error condition or failure case is being handled or checked.

    oai_token-act-pair · claude-4-5-sonnet
    Triggered by @jyhe0408

    indications of program errors or failure states in technical contexts, including error-handling, failure messages, and usage/reporting instructions.

    oai_token-act-pair · gpt-5
    Triggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-12b-pt/resid_post/layer_31_width_16k_l0_medium
    Prompts (Dashboard)
    392,802 prompts, 256 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Negative Logits
     if
    0.69
    if
    0.62
    try
    0.62
    cida
    0.61
    ร์
    0.57
    TRY
    0.56
     savory
    0.54
     headway
    0.54
    ถ้า
    0.54
     фина
    0.52
    POSITIVE LOGITS
     abort
    1.01
     break
    0.99
     exit
    0.95
    exit
    0.94
     EXIT
    0.90
    abort
    0.88
     BREAK
    0.87
    perror
    0.86
     discontinue
    0.86
     Break
    0.85
    Activations Density 0.013%

    No Known Activations