© 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
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 124152
    Prev
    Next
    INDEX
    Explanations

    punctuation and foreign words

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

    The neuron is detecting punctuation marks (commas, dashes, parentheses, colons, question marks, etc.) within the text.

    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
     wykonane
    -1.59
    parfum
    -1.59
    spod
    -1.57
    watercolor
    -1.49
     mocking
    -1.47
     silencioso
    -1.46
     barvy
    -1.45
    numerus
    -1.45
     willkommen
    -1.42
     tourist
    -1.40
    POSITIVE LOGITS
     to
    2.52
     but
    1.70
    whether
    1.48
     cabine
    1.48
    aries
    1.38
     щоб
    1.36
    桖
    1.35
    geddon
    1.35
    that
    1.34
     как
    1.34
    Activations Density 0.037%

    No Known Activations