© 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. 120085
    Prev
    Next
    INDEX
    Explanations

    same, similar

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

    This neuron activates on words that express similarity or identity (e.g. “same,” “similar,” “mesmo,” “exact,” “associated”).

    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
     eigentliche
    -2.09
    釭
    -2.09
    ated
    -2.03
     gesamte
    -1.97
     Государ
    -1.97
    蚡
    -1.96
     to
    -1.93
    乆
    -1.87
    ？！”
    -1.84
    🢒
    -1.77
    POSITIVE LOGITS
     favoritas
    1.92
     —
    1.91
    焢
    1.83
     exposes
    1.73
     Medit
    1.70
    釹
    1.69
    製の
    1.69
     idiota
    1.68
     Ouverture
    1.68
     we
    1.67
    Activations Density 0.005%

    No Known Activations