© 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. 22-GEMMASCOPE-RES-131K
    7. 66774
    Prev
    Next
    INDEX
    Explanations

    everything/both/neither

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

    The neuron generally picks out salient content words—especially technical terms, key nouns, and important descriptors.

    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
     other
    -0.98
     or
    -0.90
    riguez
    -0.90
    屹
    -0.88
     among
    -0.85
    矗
    -0.84
     različ
    -0.84
     because
    -0.81
     between
    -0.80
    ər
    -0.79
    POSITIVE LOGITS
     både
    1.55
    ทั้ง
    1.27
    everything
    1.23
     everything
    1.20
    Everything
    1.14
    both
    1.11
     weder
    1.09
    BOTH
    1.09
     EVERYTHING
    1.09
    各项
    1.06
    Activations Density 0.258%

    No Known Activations