© 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-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 10018
    Prev
    Next
    INDEX
    Explanations

    relative pronouns which/who

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

    The neuron selectively activates on numeric tokens—especially decimal‐formatted statistics and measurements—i.e. it finds floating‐point numbers and other numeric data.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-12b-pt/resid_post/layer_31_width_16k_l0_medium
    Prompts (Dashboard)
    392,802 prompts, 256 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     territorial
    0.57
    তথ্য
    0.56
    都不会
    0.54
     kinetic
    0.53
    都不
    0.52
    挪
    0.52
    那天
    0.52
    तीर्ण
    0.50
    ACTIVITY
    0.50
    潵
    0.50
    POSITIVE LOGITS
     која
    0.68
     koja
    0.66
     indicada
    0.64
     koju
    0.63
    ందని
    0.63
     considerada
    0.62
     решила
    0.60
     оказалась
    0.60
     якая
    0.59
     كبيرة
    0.57
    Activations Density 0.027%

    No Known Activations