© 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. 7156
    Prev
    Next
    INDEX
    Explanations

    pledge

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

    This neuron detects the first non‐zero (i.e. first content) token at the start of each new text segment or paragraph.

    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
     întreb
    0.80
     rookie
    0.78
     Jiangsu
    0.78
     답
    0.75
    の状態
    0.73
     enumer
    0.72
     outfielder
    0.72
     Hoon
    0.71
     Djokovic
    0.71
     tributary
    0.70
    POSITIVE LOGITS
    ט
    0.98
    什么
    0.86
    на
    0.82
    
    0.80
    σης
    0.78
    oing
    0.75
    טן
    0.74
    icleta
    0.73
    asie
    0.73
    <0x80>
    0.72
    Activations Density 0.000%

    No Known Activations