© 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. 42643
    Prev
    Next
    INDEX
    Explanations

    auxiliary verbs and common verbs

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

    The neuron detects tokens composed entirely of uppercase letters (shouty all-caps words or acronyms).

    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
     our
    -1.30
    Istorija
    -1.25
    timbangkan
    -1.21
     Leidenschaft
    -1.18
     "}";
    -1.16
    んだよね
    -1.14
     włosy
    -1.13
     nostre
    -1.10
    顏色
    -1.08
     vêtement
    -1.08
    POSITIVE LOGITS
     so
    1.31
     for
    1.24
    夜
    1.06
    任何
    1.02
    with
    1.02
    couldn
    1.02
    更何况
    1.01
     Cle
    0.99
    机
    0.99
    Mu
    0.98
    Activations Density 0.082%

    No Known Activations