© 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. 90612
    Prev
    Next
    INDEX
    Explanations

    newline and delimiters

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

    The neuron fires on explicit escaped whitespace sequences—i.e. the “\t” (tab) and “\n” (newline) tokens.

    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
    lindo
    -1.80
     sebelum
    -1.70
     厂家
    -1.67
    蛩
    -1.66
    
    -1.65
    iliki
    -1.58
     ギャラリー
    -1.55
     attiva
    -1.54
     each
    -1.54
     their
    -1.54
    POSITIVE LOGITS
     andere
    1.82
     jiné
    1.64
     annat
    1.64
    ='
    1.53
     інші
    1.52
    +
    1.51
     entiendo
    1.48
    とにかく
    1.46
     pienso
    1.42
    lerde
    1.41
    Activations Density 0.012%

    No Known Activations