© 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. 10328
    Prev
    Next
    INDEX
    Explanations

    even though

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

    The neuron responds strongly to concessive discourse markers, especially the word “even” in phrases like “even though.”

    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
     want
    -1.91
    );
    -1.77
     more
    -1.73
    )
    -1.67
     Maybe
    -1.66
     but
    -1.64
     new
    -1.64
     insbesondere
    -1.63
    Therefore
    -1.61
     presumably
    -1.60
    POSITIVE LOGITS
     nimmer
    1.88
     verhind
    1.78
    sometimes
    1.64
     leaſt
    1.60
     dichter
    1.55
    iments
    1.49
     verhe
    1.49
     heilig
    1.48
    becoming
    1.46
    鱚
    1.46
    Activations Density 0.011%

    No Known Activations