© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 66825
    Prev
    Next
    INDEX
    Explanations

    not so

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

    The neuron activates on negation–intensifier constructions, in particular the phrase “not so” (and variants like “not too”).

    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
     any
    -2.06
     without
    -1.99
     this
    -1.89
     or
    -1.84
     when
    -1.83
    ;
    -1.77
     such
    -1.73
     it
    -1.70
     should
    -1.67
    滙
    -1.62
    POSITIVE LOGITS
    ”
    1.69
    ’’
    1.59
     TextAlign
    1.55
     technologically
    1.54
     interessante
    1.54
     elegante
    1.51
     chaleco
    1.50
    蹰
    1.49
    vantaged
    1.48
     sebel
    1.45
    Activations Density 0.012%

    No Known Activations