© 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. 34-GEMMASCOPE-RES-131K
    7. 64095
    Prev
    Next
    INDEX
    Explanations

    rare or scarce

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

    The neuron is primarily detecting forms of the copula “to be” (e.g. “is,” “are”).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
    зыв
    -0.90
    thma
    -0.90
    >−
    -0.86
    otherapy
    -0.84
    ebly
    -0.83
    vertible
    -0.82
     obvio
    -0.80
    oprene
    -0.80
    lassi
    -0.80
    ndash
    -0.80
    POSITIVE LOGITS
     rare
    4.38
     rarity
    4.16
     scarce
    3.72
     rarer
    3.58
     scarcity
    3.53
     rarest
    3.50
     Rare
    3.22
    Rare
    3.11
     RARE
    2.97
    rare
    2.86
    Activations Density 0.074%

    No Known Activations