© 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. 4691
    Prev
    Next
    INDEX
    Explanations

    applications across fields

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

    The neuron activates on mentions of practical “applications” or use‐case domains, especially where a technology or material is described as being applied (e.g. “in biomedical applications”).

    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
     What
    0.79
     antics
    0.79
     what
    0.74
     Mayor
    0.73
     prostitutes
    0.73
     wages
    0.72
     reigning
    0.72
    隍
    0.72
     Tyler
    0.72
     Gayle
    0.71
    POSITIVE LOGITS
     discretized
    0.93
     Anwendungen
    0.91
     perturbative
    0.89
     inhomogeneous
    0.88
     anisotropic
    0.87
     informatics
    0.87
    isotropic
    0.86
    middlewares
    0.86
    ochastic
    0.83
     astrophysical
    0.82
    Activations Density 0.036%

    No Known Activations