© 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. 12023
    Prev
    Next
    INDEX
    Explanations

    performance, alternatives, privacy, open

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

    The neuron activates on promotional comparative and superlative language that highlights superior performance or quality by comparing a product to others.

    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
    地点
    0.58
    곳
    0.57
     местах
    0.57
    地點
    0.54
    Reuse
    0.54
    Located
    0.54
    जहां
    0.54
     বিশাল
    0.53
    𝙲
    0.53
     제조업
    0.52
    POSITIVE LOGITS
     alternatives
    1.42
     alternative
    1.41
     alternativa
    1.35
     solution
    1.31
    alternative
    1.24
     альтерна
    1.23
     successor
    1.18
     alternativ
    1.17
     tool
    1.14
     alternatif
    1.14
    Activations Density 0.453%

    No Known Activations