© 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. 98623
    Prev
    Next
    INDEX
    Explanations

    unexpected circumstances

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

    This neuron strongly activates on occurrences of “unexpected” (and its morphological variants).

    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
     to
    -1.88
     trovo
    -1.67
    actéristi
    -1.66
     You
    -1.60
     député
    -1.57
    Aiheesta
    -1.56
     estimés
    -1.56
     giovane
    -1.54
     jeweiligen
    -1.54
    ра
    -1.53
    POSITIVE LOGITS
     but
    2.28
    .
    1.81
    ness
    1.78
     things
    1.58
     puts
    1.54
    というか
    1.54
     circumstances
    1.51
     situations
    1.49
     occurrence
    1.48
    <i>
    1.43
    Activations Density 0.014%

    No Known Activations