© 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. 22-GEMMASCOPE-RES-131K
    7. 34277
    Prev
    Next
    INDEX
    Explanations

    completion of events

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

    The neuron strongly activates on statements announcing that something is completed or wrapped up (e.g. “wrapped up,” “is now finished,” “are over again,” “are closed”).

    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
    })*/
    -0.94
    Tại
    -0.88
    Πηγή
    -0.87
     rst
    -0.86
    eventually
    -0.85
     primeras
    -0.85
     условиях
    -0.84
    Mura
    -0.82
    Màu
    -0.82
    pregunt
    -0.82
    POSITIVE LOGITS
     now
    1.60
     Теперь
    1.32
     next
    1.30
     теперь
    1.21
    あとは
    1.20
    Теперь
    1.11
    Now
    1.09
    只剩下
    1.07
    Next
    1.07
    接著
    1.04
    Activations Density 0.076%

    No Known Activations