© 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. 10-GEMMASCOPE-RES-131K
    7. 1661
    Prev
    Next
    INDEX
    Explanations

    evacuation and evacuations

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

    The neuron activates on various forms of the root “evacu-” (evacuate, evacuated, evacuating, evacuation, evacuations).

    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
    ar
    -2.69
    C
    -2.69
    V
    -2.33
    P
    -2.33
    en
    -2.23
    G
    -2.22
    H
    -2.20
    犼
    -2.14
    W
    -2.14
     пары
    -2.14
    POSITIVE LOGITS
    鶿
    2.33
    栳
    2.28
     adil
    2.20
    妡
    2.20
    𞤥
    2.19
     ravi
    2.16
     luto
    2.16
     hermos
    2.16
     umo
    2.14
     orar
    2.09
    Activations Density 0.004%

    No Known Activations