© 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-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 2093
    Prev
    Next
    INDEX
    Explanations

    Okay, let's break down

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

    This neuron detects numerical tokens—especially floating‐point numbers—used for measurements, citations, or reference values.

    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
     apricots
    0.59
     Bosco
    0.58
     butterflies
    0.58
     chern
    0.58
    cherche
    0.57
    AppBsky
    0.57
     Kotlin
    0.57
    甑
    0.57
    ⚈
    0.57
     एव
    0.55
    POSITIVE LOGITS
    iba
    0.54
     भूल
    0.50
    となります
    0.48
    damping
    0.47
    Needs
    0.46
    荼
    0.46
     تاز
    0.45
    Contours
    0.45
    تمل
    0.44
     نشده
    0.44
    Activations Density 0.056%

    No Known Activations