© 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. 59183
    Prev
    Next
    INDEX
    Explanations

    meta tags and headers

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

    The main thing this neuron does is detect the English definite article “the” (in any capitalization).

    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
     covid
    -1.13
    Covid
    -1.00
    <0xF4>
    -0.99
    РО
    -0.98
     Covid
    -0.94
    клу
    -0.94
    COVID
    -0.93
    anadium
    -0.92
     Shall
    -0.91
    {$
    -0.91
    POSITIVE LOGITS
     llego
    0.93
     apparently
    0.86
    Horario
    0.84
     certainly
    0.82
     estaría
    0.82
     slamming
    0.81
     dozens
    0.81
     did
    0.79
     permitirá
    0.79
     respetar
    0.78
    Activations Density 0.002%

    No Known Activations