© 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. 87168
    Prev
    Next
    INDEX
    Explanations

    orientation and flipping

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

    The main thing this neuron does is detect words describing flipping or orientation changes (e.g. overturn, flip, upside-down, upright).

    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
    ↵↵↵
    -2.06
    >
    -1.96
    0
    -1.88
    {
    -1.77
    6
    -1.70
    2
    -1.69
    op
    -1.63
     destes
    -1.61
    5
    -1.59
    o
    -1.55
    POSITIVE LOGITS
    
    2.08
    婓
    1.64
     fundamentos
    1.64
     publicados
    1.52
     pedidos
    1.52
     kämp
    1.50
     utilisez
    1.49
     fundación
    1.48
     tragé
    1.47
     gyer
    1.45
    Activations Density 0.036%

    No Known Activations