© 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. 10842
    Prev
    Next
    INDEX
    Explanations

    commands to add or change state

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

    The neuron fires on prominent action verbs—especially instruction or update verbs (e.g. added, published, select, click, integrate)—marking key procedural or change‐of‐state steps.

    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
    有时
    0.68
    然而
    0.63
    便于
    0.63
    鬻
    0.63
    Vertical
    0.62
    Managing
    0.62
    শরীর
    0.60
    ORN
    0.59
    우리
    0.59
    衙
    0.59
    POSITIVE LOGITS
     č
    0.84
     Č
    0.82
     zwei
    0.81
     vitamina
    0.77
     isos
    0.77
     siete
    0.74
     deset
    0.73
     drei
    0.70
     raincoat
    0.70
     cytokinin
    0.70
    Activations Density 0.248%

    No Known Activations