© 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-3-27B-IT
    5. 31-GEMMASCOPE-2-RES-262K
    7. 3503
    Prev
    Next
    INDEX
    Explanations

    numbers and starting phrases

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

    sentences containing numeric quantities or math/problem-statement content (numbers, calculations, or math-related tokens).

    oai_token-act-pair · gpt-5-miniTriggered by @vetterc0
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-27b-it/resid_post/layer_31_width_262k_l0_medium
    Prompts (Dashboard)
    238,145 prompts, 512 tokens each
    Dataset (Dashboard)
    lmsys + oasst1
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    融资
    0.39
    权威
    0.39
    级别
    0.39
    贯彻
    0.38
     bolj
    0.38
    还得
    0.38
     financiación
    0.37
    뀔
    0.37
     કરવાની
    0.37
    这也
    0.37
    POSITIVE LOGITS
     A
    0.46
     நான்கு
    0.46
    _{
    0.45
     $(
    0.44
     three
    0.44
     $=
    0.42
     ある
    0.42
    A
    0.42
     identical
    0.42
     तीन
    0.42
    Activations Density 0.131%

    No Known Activations