© 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
    3. Gemma-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 8622
    Prev
    Next
    INDEX
    Explanations

    `local` optimization

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

    The neuron fires on isolated, content-bearing words that act as category labels—e.g. single nouns or adjectives like “blue,” “string,” or “pattern.”

    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
     дохода
    0.93
     escolares
    0.89
    蕌
    0.86
     moradores
    0.85
    铁路
    0.85
     ఆదాయ
    0.84
    萂
    0.84
     taxed
    0.84
     सेंट्रल
    0.83
     inflación
    0.82
    POSITIVE LOGITS
    al
    0.89
    ↵
    0.87
    s
    0.82
    m
    0.77
    0.73
    TITLE
    0.72
    Pro
    0.70
    om
    0.70
    0
    0.68
    Sp
    0.68
    Activations Density 0.000%

    No Known Activations