© 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. 22-GEMMASCOPE-RES-131K
    4. 66669
    Prev
    Next
    INDEX
    Explanations

    marking content boundaries

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

    The neuron fires on inline punctuation and symbols (parentheses, ellipses, asterisks, dashes, etc.) that mark asides or list-style formatting.

    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
    compañ
    -1.00
    hooting
    -0.99
    出した
    -0.92
    ologists
    -0.91
    bucks
    -0.89
    Shooting
    -0.89
    demic
    -0.85
    intenance
    -0.85
    UNGEN
    -0.84
    REEN
    -0.84
    POSITIVE LOGITS
     again
    1.14
     because
    1.09
     especially
    1.02
     certainly
    0.98
     many
    0.97
     depends
    0.96
     huge
    0.95
    これも
    0.94
     different
    0.94
     опять
    0.92
    Activations Density 0.014%

    No Known Activations