© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 31650
    Prev
    Next
    INDEX
    Explanations

    whale watching

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

    The neuron fires on mentions of whales (e.g. “whale watching,” “whales returning”), i.e. any reference to whales.

    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
     Aguilera
    -0.77
    unless
    -0.70
     dire
    -0.69
    athione
    -0.67
     Goats
    -0.65
    default
    -0.65
     Manganese
    -0.64
    seal
    -0.64
     Outlets
    -0.64
     ROY
    -0.64
    POSITIVE LOGITS
     whale
    1.46
     whales
    1.25
    🐋
    1.15
     Whale
    1.14
    whale
    1.09
    🐳
    1.08
    Whale
    1.04
     Whales
    0.92
     migrating
    0.87
     wh
    0.85
    Activations Density 0.013%

    No Known Activations