© 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. 23860
    Prev
    Next
    INDEX
    Explanations

    things of the past, fading

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

    The neuron fires on idiomatic “of the” constructions—e.g. “fall to the wayside,” “thing of the past,” and “stand the test of time.”

    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
     your
    -1.49
    ことができました
    -1.32
     said
    -1.32
    していて
    -1.29
    kör
    -1.27
    聒
    -1.24
    from
    -1.22
     and
    -1.21
    分に
    -1.20
     ensures
    -1.18
    POSITIVE LOGITS
     condomínio
    1.39
    ৄ
    1.34
    weite
    1.34
     birkaç
    1.34
     juntamente
    1.33
     reconsider
    1.33
     dotyczą
    1.30
    高
    1.30
    ationen
    1.29
    richtungen
    1.29
    Activations Density 0.025%

    No Known Activations