© 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. 22-GEMMASCOPE-RES-131K
    7. 73693
    Prev
    Next
    INDEX
    Explanations

    question-answer pairs, calculations

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

    The neuron fires on punctuation‐heavy separator tokens (e.g. commas, arrows “->” or “=>”, quotes with trailing commas) that mark value or example boundaries in code/test‐case listings.

    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
     this
    -1.16
     of
    -1.11
    useNewUrlParser
    -0.98
    藹
    -0.97
    できるので
    -0.91
     světě
    -0.89
    krankungen
    -0.88
     ngunit
    -0.88
    entado
    -0.88
     மற்ற
    -0.88
    POSITIVE LOGITS
     donde
    1.01
    ̀nh
    0.99
     princípio
    0.99
     flera
    0.95
     tendremos
    0.93
     sería
    0.91
     começo
    0.90
     fluxo
    0.90
     bloco
    0.89
     también
    0.88
    Activations Density 0.052%

    No Known Activations