© 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. 11872
    Prev
    Next
    INDEX
    Explanations

    "eat the frog"

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

    This neuron specializes in spotting formal legal or regulatory jargon and reference phrases (e.g. “established from time to time,” “third party,” “cooperate”).

    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
    之路
    0.94
    女性
    0.89
    唑
    0.88
    
    0.88
    アイドル
    0.87
    我就
    0.86
    𝙡
    0.85
    𝑙
    0.85
     aspetto
    0.84
     arrête
    0.84
    POSITIVE LOGITS
     consistent
    0.78
    eps
    0.71
     EPS
    0.71
    etermined
    0.69
     AIP
    0.69
     Edison
    0.68
     inflated
    0.68
     convoluted
    0.67
     জনে
    0.67
     тыс
    0.66
    Activations Density 0.003%

    No Known Activations