© 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. 116480
    Prev
    Next
    INDEX
    Explanations

    runway, taxiing, airport operations

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

    The neuron primarily activates on numeric tokens and numeric measurements (e.g. runway lengths, counts, or other digit sequences).

    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
    Obrázky
    -0.81
     cuck
    -0.78
    oniec
    -0.75
    jini
    -0.74
    hejiang
    -0.73
    扎
    -0.72
     forester
    -0.71
     Ape
    -0.70
    プレス
    -0.70
    DebuggerNonUser
    -0.69
    POSITIVE LOGITS
     runway
    1.17
     taxi
    0.91
     Runway
    0.90
    LEFT
    0.81
     LEFT
    0.79
     piano
    0.76
     direction
    0.76
     Direction
    0.76
     pavement
    0.75
     Tonnen
    0.75
    Activations Density 0.027%

    No Known Activations