© 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. 10-GEMMASCOPE-RES-131K
    7. 8072
    Prev
    Next
    INDEX
    Explanations

    in ordering instructions

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

    The neuron consistently fires on the standalone preposition “in,” especially in mathematical statements (e.g. “in R,” “in Gᵏ,” “in Ω”).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
    𐬹
    -1.70
     was
    -1.64
    -1.61
    侂
    -1.59
    },
    -1.58
    ७
    -1.50
    Là
    -1.48
    鑪
    -1.47
    鋇
    -1.47
    на
    -1.45
    POSITIVE LOGITS
     the
    2.67
     our
    1.93
    慙
    1.87
     we
    1.86
     your
    1.66
     their
    1.66
    怍
    1.59
    肏
    1.55
     OF
    1.49
    俛
    1.49
    Activations Density 0.029%

    No Known Activations