© 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. 72756
    Prev
    Next
    INDEX
    Explanations

    sameness and similarity

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

    The neuron fires on words and word fragments that express sameness or close similarity (e.g. same, exactly, similar, parallel, identical).

    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
     itself
    -1.23
     its
    -1.03
     it
    -1.02
    它的
    -0.97
    因为它
    -0.93
    ამ
    -0.92
     instead
    -0.89
     sometimes
    -0.88
     was
    -0.85
     adicionais
    -0.85
    POSITIVE LOGITS
     одина
    2.78
     identical
    2.52
     differences
    2.17
    identical
    2.17
    相同
    2.03
     same
    1.94
    Differences
    1.92
     their
    1.91
     respective
    1.88
     alike
    1.88
    Activations Density 0.240%

    No Known Activations