© 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. 12436
    Prev
    Next
    INDEX
    Explanations

    multi-punctuation marks

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

    The neuron flags runs of repeated exclamation (and similar emphatic) punctuation marks.

    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
     misguided
    -1.84
     ruthless
    -1.82
    autogui
    -1.80
     brilliant
    -1.77
     Because
    -1.70
     exuberant
    -1.70
     However
    -1.70
     ?
    -1.66
     ideia
    -1.65
     poignant
    -1.64
    POSITIVE LOGITS
     forbindelse
    2.23
    木坂
    2.16
    ’
    1.98
    ’，
    1.95
     anhänger
    1.91
    whose
    1.88
    ко
    1.86
     zoude
    1.84
     dando
    1.82
     Verenig
    1.82
    Activations Density 0.008%

    No Known Activations