© 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. 24-GEMMASCOPE-2-RES-16K
    7. 7908
    Prev
    Next
    INDEX
    Explanations

    dependencies, power, dosage, balance

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

    The neuron activates on abstract nouns that denote positive or beneficial concepts, services, or improvements (e.g., life, quality, errands, improvement).

    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
    
    0.79
    ፥
    0.72
     Рас
    0.71
     ছয়
    0.66
    ﹗
    0.65
    ﹍
    0.65
    icolor
    0.64
    楽しく
    0.64
    Из
    0.63
    ․
    0.63
    POSITIVE LOGITS
     Maurizio
    0.74
     když
    0.71
     wzglę
    0.71
     despite
    0.70
     cuando
    0.69
     lorsqu
    0.69
     logrado
    0.69
     ettiği
    0.69
     when
    0.68
     perovsk
    0.68
    Activations Density 0.507%

    No Known Activations