© 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. 22-GEMMASCOPE-RES-131K
    7. 65541
    Prev
    Next
    INDEX
    Explanations

    above, same

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

    The neuron detects occurrences of the phrase “same as above,” i.e. self-referential cross-references in text.

    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
    atrici
    -1.04
    τού
    -0.98
     stolet
    -0.98
     برخی
    -0.96
     इस
    -0.94
     piese
    -0.93
    runnable
    -0.92
    Buku
    -0.91
    Симпто
    -0.91
    erski
    -0.90
    POSITIVE LOGITS
    </strong>
    1.23
    日発売
    1.09
    .,
    0.97
     स्क्रीन
    0.96
    .
    0.94
     mia
    0.93
    $
    0.93
     ANG
    0.91
    棕色
    0.91
     honours
    0.90
    Activations Density 0.006%

    No Known Activations