© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 113767
    Prev
    Next
    INDEX
    Explanations

    airplane crashes

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

    This neuron activates on mentions of aircraft—words like “airplane,” “plane,” “jet,” or related aviation terms.

    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
    苼
    -2.27
    </b>
    -1.93
     unfor
    -1.89
    锇
    -1.84
    饺
    -1.83
     enorme
    -1.81
    características
    -1.79
     малень
    -1.75
    bzw
    -1.75
    ۣ
    -1.74
    POSITIVE LOGITS
    </h3>
    2.19
     arbeitete
    2.17
    猻
    1.97
     schlechte
    1.89
     With
    1.88
     mehrfach
    1.84
     zahlreiche
    1.81
     längere
    1.80
     ($("#
    1.78
     erhielt
    1.75
    Activations Density 0.008%

    No Known Activations