© 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. 10-GEMMASCOPE-RES-131K
    7. 113123
    Prev
    Next
    INDEX
    Explanations

    negation and inadequacy

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

    The neuron strongly activates on negation tokens—especially the “n't” ending and words like “no” or “not.”

    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
    ниях
    -1.54
     unſ
    -1.52
    Merkmale
    -1.51
     leaſt
    -1.49
    ｐ
    -1.48
     skak
    -1.45
     minimalis
    -1.44
     graden
    -1.41
     geograf
    -1.41
    polate
    -1.41
    POSITIVE LOGITS
    ↵
    1.77
     their
    1.44
    '
    1.40
     time
    1.40
     indruk
    1.37
    1
    1.29
     gekomen
    1.27
     people
    1.24
     or
    1.23
     vroeg
    1.23
    Activations Density 0.026%

    No Known Activations