© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 12641
    Prev
    Next
    INDEX
    Explanations

    GitHub organization or assignment

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

    The neuron is highly sensitive to small function‐word tokens that form part of phrasal‐verb constructions (e.g. “attach to,” “add up to,” “whether requested to,” etc.).

    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
    ر
    0.85
     pert
    0.77
        
    0.72
    راب
    0.71
     whereupon
    0.70
     deprec
    0.69
     inverter
    0.69
    ўні
    0.69
    あり
    0.68
     epitaxial
    0.68
    POSITIVE LOGITS
     Precio
    0.86
    स्टोन
    0.82
    पाउंड
    0.78
    ího
    0.78
    的有
    0.78
     Среди
    0.77
    प्लीट
    0.77
     Đấy
    0.76
     Año
    0.75
    फॉर्मेशन
    0.75
    Activations Density 0.000%

    No Known Activations