© 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. 34-GEMMASCOPE-RES-131K
    7. 48815
    Prev
    Next
    INDEX
    Explanations

    head-scratching confusion

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

    The neuron activates on variants of “scratch(ing) your head” (e.g. “head-scratching,” “scratching my head”)—idioms expressing puzzlement.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
    龇
    -0.94
    urie
    -0.93
     gruesome
    -0.90
    咂
    -0.87
    食感
    -0.87
     negotiations
    -0.83
    OMET
    -0.82
     motiv
    -0.82
    っているので
    -0.79
     cruelly
    -0.78
    POSITIVE LOGITS
     scratching
    4.50
     scratch
    3.92
    scratch
    3.84
     scratched
    3.53
     Scratch
    3.39
    Scratch
    3.33
     scratches
    2.72
    挠
    1.75
     scrat
    1.72
     puzzled
    1.52
    Activations Density 0.068%

    No Known Activations