© 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. 10-GEMMASCOPE-RES-131K
    7. 58997
    Prev
    Next
    INDEX
    Explanations

    peninsulas and related geography

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

    The neuron activates on mentions of peninsulas (the word “peninsula” itself or proper names of specific peninsulas).

    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
    𞤢
    -1.79
     a
    -1.74
    塒
    -1.64
     Ahnung
    -1.63
     in
    -1.60
     talaga
    -1.51
    IonicModule
    -1.44
    >
    -1.42
     this
    -1.40
    та
    -1.39
    POSITIVE LOGITS
    蓠
    1.89
     några
    1.68
     зданий
    1.55
     cuyas
    1.54
     letzt
    1.46
    losti
    1.45
     \
    1.45
    delaire
    1.44
     suivi
    1.41
     flaky
    1.41
    Activations Density 0.025%

    No Known Activations