© 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. 43119
    Prev
    Next
    INDEX
    Explanations

    playing around with

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

    This neuron detects the self-referential phrase “playing around with,” marking when the text says the author is experimenting with something.

    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
     Those
    -1.78
     umana
    -1.69
     Rather
    -1.66
     Neither
    -1.64
     Instead
    -1.58
    &
    -1.57
    %
    -1.50
    那些
    -1.49
    =
    -1.47
    事實
    -1.44
    POSITIVE LOGITS
     różnych
    1.66
    퐸
    1.64
     několik
    1.61
     aprendido
    1.59
     ways
    1.59
    𝔈
    1.58
    してみて
    1.57
     také
    1.57
     zlep
    1.55
     également
    1.55
    Activations Density 0.016%

    No Known Activations