© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 71111
    Prev
    Next
    INDEX
    Explanations

    numbers and their units

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

    The neuron is picking out numeric quantity expressions (e.g. counts, doses, durations, measurements) in scientific 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
    的不同
    -0.85
    RECIP
    -0.82
     even
    -0.81
    Bukan
    -0.81
    ikasi
    -0.80
    一些
    -0.78
    ijnt
    -0.78
    čenie
    -0.78
     कुछ
    -0.77
    ὑ
    -0.77
    POSITIVE LOGITS
     their
    1.04
     each
    0.91
    yaw
    0.88
    each
    0.85
     leurs
    0.84
    rück
    0.83
    互联网档案馆
    0.82
    同一
    0.80
    なるべく
    0.79
     jeweils
    0.79
    Activations Density 0.195%

    No Known Activations