© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 69900
    Prev
    Next
    INDEX
    Explanations

    words ending in ed or ized

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

    The neuron activates on words ending in the “-ed” suffix (including “-ized”) – i.e. past-tense or past-participle/adjectival forms.

    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
    ).
    -3.23
    ↵
    -3.11
    ”:
    -3.05
    了下
    -2.94
    ):
    -2.84
    Ꭳ
    -2.84
    ”;
    -2.77
    ⟞
    -2.69
    随即
    -2.67
    当即
    -2.66
    POSITIVE LOGITS
    ⑉
    2.64
    鬭
    2.64
     articoli
    2.59
    ugeot
    2.47
    纽约
    2.45
    nambuco
    2.45
    的
    2.44
     pernikahan
    2.41
     jabatan
    2.41
     The
    2.39
    Activations Density 0.111%

    No Known Activations