© 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. 22-GEMMASCOPE-RES-131K
    7. 34770
    Prev
    Next
    INDEX
    Explanations

    programming keywords and variable names

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

    The neuron is detecting numeric tokens—i.e. standalone numbers or digits—in code snippets.

    oai_token-act-pair · o4-miniTriggered 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
    Not in Any Lists

    No Comments

    Negative Logits
    \'{
    -0.93
    èm
    -0.88
    激しい
    -0.83
     ornamented
    -0.82
    会不会
    -0.82
    吃了
    -0.81
    mostly
    -0.81
    県道
    -0.81
     subsequent
    -0.80
    Tempor
    -0.80
    POSITIVE LOGITS
     марта
    0.98
     августа
    0.93
    addError
    0.90
    //!
    0.89
    month
    0.87
    ibilidade
    0.86
    vaded
    0.86
    gangss
    0.86
    ()-
    0.85
    animator
    0.84
    Activations Density 0.023%

    No Known Activations