© 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. 11035
    Prev
    Next
    INDEX
    Explanations

    The neuron detects Biblical or proverb‐style quotations, especially the “let those without sin cast the first stone”‐type allusions.

    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
    铤
    0.66
    违
    0.55
    negoti
    0.54
    pflicht
    0.54
    constraintTop
    0.54
    違
    0.54
     calming
    0.53
    谈判
    0.53
    闯
    0.53
    违法
    0.53
    POSITIVE LOGITS
     slander
    1.77
    诋
    1.63
    詆
    1.63
     accusations
    1.59
     accusing
    1.58
     criticisms
    1.57
     accusation
    1.56
     criticizing
    1.55
     criticism
    1.55
    誹
    1.55
    Activations Density 0.456%

    No Known Activations