© 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. 35002
    Prev
    Next
    INDEX
    Explanations

    colors red green blue

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

    The neuron activates on color words (e.g., “green,” “red,” “blue”).

    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
    给大家
    -0.74
     readiness
    -0.68
     eagerness
    -0.67
     руке
    -0.67
    Furthermore
    -0.66
     maxWidth
    -0.65
    ்ச
    -0.65
     του
    -0.65
    cellulose
    -0.65
     عبد
    -0.64
    POSITIVE LOGITS
     color
    0.93
     colour
    0.93
    mailing
    0.86
     strips
    0.83
     signal
    0.80
     spots
    0.80
    alist
    0.77
    碕
    0.77
     Signal
    0.77
    ideae
    0.73
    Activations Density 0.042%

    No Known Activations