© 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. 34-GEMMASCOPE-RES-131K
    7. 60197
    Prev
    Next
    INDEX
    Explanations

    vertical align center

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

    This neuron fires on CSS declarations—specifically the punctuation, colons, semicolons, numeric values, and unit symbols (%) that mark property‐value pairs in style rules.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
     gap
    -0.98
    marginBottom
    -0.87
    ugno
    -0.81
     marginBottom
    -0.79
    voř
    -0.76
    ಞ
    -0.74
    デーション
    -0.74
     Gap
    -0.73
    asının
    -0.73
     žád
    -0.71
    POSITIVE LOGITS
    centerY
    1.41
     vertical
    1.13
     middle
    1.12
    middle
    1.09
    Center
    1.02
     CENTER
    1.02
     center
    1.02
    垂直
    1.00
    center
    1.00
    CENTER
    0.94
    Activations Density 0.026%

    No Known Activations