© 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. 22-GEMMASCOPE-RES-131K
    7. 130317
    Prev
    Next
    INDEX
    Explanations

    The neuron is selectively detecting the token “bed” (and its immediate variants) wherever it appears in the text.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    bed and things that follow

    np_acts-logits-general · gemini-2.5-flash-lite
    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
     he
    -3.03
    鼱
    -2.83
    并且
    -2.73
    และ
    -2.50
    </td>
    -2.45
    峮
    -2.41
    垍
    -2.38
    必要がある
    -2.34
    ℣
    -2.30
    必要があります
    -2.23
    POSITIVE LOGITS
    These
    2.58
    after
    2.55
    ﮟ
    2.52
     What
    2.34
     staunch
    2.31
     This
    2.30
    entikan
    2.28
     енер
    2.27
    Both
    2.22
     dengan
    2.20
    Activations Density 0.007%

    No Known Activations