© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 15345
    Prev
    Next
    INDEX
    Explanations

    so it was

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

    The neuron is triggered by occurrences of the word “so” used as a discourse marker or intensifier.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
    に
    -2.06
    があり
    -2.02
    ですが
    -2.02
    と
    -2.02
    そして
    -1.83
    <bos>
    -1.77
    がある
    -1.70
    がありますが
    -1.69
    や
    -1.65
    は
    -1.61
    POSITIVE LOGITS
    為に
    2.19
     that
    1.92
    ができます
    1.79
    感じです
    1.77
    の絵
    1.70
    来週
    1.66
    ほうがいい
    1.63
    が一番
    1.61
    の方も
    1.58
    になったら
    1.55
    Activations Density 0.038%

    No Known Activations