© 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. 14361
    Prev
    Next
    INDEX
    Explanations

    synsets, exoplanets, DSLs

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

    The neuron fires strongly whenever the text mentions someone carrying out a specific task or activity.

    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
    ﻌ
    0.98
    ﺤ
    0.95
    ﻤ
    0.92
    ిత
    0.90
    ாக
    0.89
    ﻘ
    0.88
     причины
    0.86
    ﻐ
    0.86
    ညီ
    0.84
    просы
    0.83
    POSITIVE LOGITS
     rural
    0.81
     personnel
    0.81
     demi
    0.80
     measure
    0.73
     spic
    0.71
    ן
    0.70
    rural
    0.70
     ported
    0.70
     taur
    0.70
     bele
    0.68
    Activations Density 0.000%

    No Known Activations