© 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. 119009
    Prev
    Next
    INDEX
    Explanations

    game activities

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

    The neuron fires on terminology associated with hunting and wildlife—especially the word “game” and related outdoors/habitat words like “brush,” “wildlife,” and “predators.”

    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
     baumwolle
    -1.66
    踌
    -1.66
    櫺
    -1.61
    幛
    -1.59
    一旁的
    -1.59
    蕖
    -1.56
     paradigma
    -1.55
     autres
    -1.53
     sors
    -1.52
    饧
    -1.52
    POSITIVE LOGITS
     is
    2.23
     стаття
    1.68
    这种
    1.66
    弊
    1.56
     яке
    1.56
    ۼ
    1.50
    尤其是
    1.48
     “
    1.48
    卜
    1.48
    となっています
    1.45
    Activations Density 0.011%

    No Known Activations