© 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. 10-GEMMASCOPE-RES-131K
    7. 100128
    Prev
    Next
    INDEX
    Explanations

    evaluating code strings

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

    the word "eval" in programming code contexts.

    oai_token-act-pair · claude-4-5-sonnetTriggered by @bfgcn7frf5

    The main thing this neuron does is detect uses of the “eval” function in code.

    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
    跸
    -2.73
    难怪
    -2.72
    艄
    -2.56
    腚
    -2.48
     were
    -2.44
    
    -2.42
    憷
    -2.41
    และ
    -2.38
    甭
    -2.36
    ⩥
    -2.36
    POSITIVE LOGITS
    1
    3.67
    After
    3.45
    If
    3.41
    7
    3.16
    With
    3.09
    When
    3.06
    )
    3.02
    4
    3.00
    For
    2.97
    To
    2.89
    Activations Density 0.004%

    No Known Activations