© 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. 33064
    Prev
    Next
    INDEX
    Explanations

    punctuation followed by is or it or he

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

    The neuron fires on the closing punctuation of the very first definitional sentence in an entry (i.e. the period at the end of the opening “X is a Y…” sentence).

    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
     veckor
    -1.16
     sø
    -1.04
     dagar
    -1.00
     τοῦ
    -1.00
     timmar
    -0.98
     tentang
    -0.97
     prefier
    -0.92
    }${
    -0.90
    ?")
    -0.89
    .$_
    -0.89
    POSITIVE LOGITS
     betrekking
    0.89
     geprüft
    0.88
     located
    0.85
     festgelegt
    0.84
     券
    0.82
    置く
    0.81
    Etimología
    0.80
    ्रीय
    0.80
    Crossing
    0.79
    ダウンロード
    0.79
    Activations Density 0.001%

    No Known Activations