© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 12755
    Prev
    Next
    INDEX
    Explanations

    if.

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

    The neuron selectively activates on the conditional conjunction “if.”

    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
    Är
    -1.08
     ролях
    -1.05
     ebenfalls
    -1.05
    ⿱
    -1.04
    brada
    -1.02
    abaikan
    -1.02
    銹
    -1.02
     didst
    -1.00
     лак
    -0.98
     ABBREVIATIONS
    -0.97
    POSITIVE LOGITS
     you
    2.52
     we
    2.48
     they
    1.90
     people
    1.87
     there
    1.79
     it
    1.66
     somebody
    1.62
     [
    1.58
     I
    1.56
     those
    1.51
    Activations Density 0.005%

    No Known Activations