© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Gemma-3-12B
    3. 24-GEMMASCOPE-2-RES-16K
    4. 12977
    Prev
    Next
    INDEX
    Explanations

    combinatorial

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

    The neuron tends to light up on uncommon or content‐heavy tokens—rare words, proper names, technical terms, and other low-frequency strings.

    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_24_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
    insert
    0.80
    information
    0.80
    genuine
    0.79
    ↵
    0.78
    lower
    0.77
    вт
    0.77
     
    0.76
    נ
    0.76
    pt
    0.75
    nb
    0.73
    POSITIVE LOGITS
     precludes
    0.95
     carinatis
    0.93
     operativos
    0.91
     dwind
    0.91
     bloquear
    0.90
    ㈠
    0.88
     preclude
    0.86
     afectadas
    0.85
     deslig
    0.84
     drows
    0.84
    Activations Density 0.000%

    No Known Activations