© 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. 11789
    Prev
    Next
    INDEX
    Explanations

    words ending in -ity, -ion, -ed

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

    The neuron specializes in spotting common English word‐ending suffixes (e.g. –tion, –ity, –ded, –tty, –fing, –trading).

    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
     I
    -1.66
    0
    -1.59
    ↵
    -1.49
     The
    -1.48
    t
    -1.37
    があります
    -1.32
     lakini
    -1.27
    だけど
    -1.26
    1
    -1.26
     imprend
    -1.23
    POSITIVE LOGITS
     宣传
    1.69
     nič
    1.60
    ⟞
    1.53
     級
    1.50
    ！！”
    1.48
    豺
    1.48
    ྜ
    1.45
    ）（
    1.45
    宴会
    1.44
    ltä
    1.44
    Activations Density 0.033%

    No Known Activations