© 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
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 111142
    Prev
    Next
    INDEX
    Explanations

    athlete

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

    The neuron selectively activates on the word “athlete” (in any casing or as part of a longer token).

    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
    ”
    -2.83
    ].
    -2.59
     scrutin
    -2.36
     affluent
    -2.36
    ."
    -2.20
    ’
    -2.20
    絎
    -2.16
     extravagant
    -2.06
    .“
    -1.99
     corrobor
    -1.98
    POSITIVE LOGITS
    峮
    3.02
    菹
    2.86
    圌
    2.53
     BRARY
    2.48
     også
    2.42
     получил
    2.30
    ꦠ
    2.30
    ſelf
    2.30
    蹌
    2.28
     extraordinaire
    2.25
    Activations Density 0.003%

    No Known Activations