© 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. 82425
    Prev
    Next
    INDEX
    Explanations

    aerial views of birds

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

    The neuron detects language describing visual vantage or perspective—especially aerial or high‐definition viewpoints (e.g. “shot … from a helicopter,” “offers an unparalleled view,” “overlooking”).

    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
    Ѓ
    -0.81
     Veld
    -0.79
    Navbar
    -0.78
     ört
    -0.75
     tourner
    -0.73
    Халык
    -0.70
    انون
    -0.70
     میز
    -0.69
    Viited
    -0.69
    GetUser
    -0.69
    POSITIVE LOGITS
     bird
    3.95
     birds
    3.50
    bird
    3.03
     Bird
    2.77
    birds
    2.72
     Birds
    2.72
    Bird
    2.66
    Birds
    2.66
     overview
    2.55
     aerial
    2.50
    Activations Density 0.047%

    No Known Activations