© 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. 15805
    Prev
    Next
    INDEX
    Explanations

    glass, vertex, filename, bowl, whisk, funding, classic

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

    The neuron fires on isolated, content‐bearing words or short labels (like “report,” “order,” “do,” “gap,” “p”) that often serve as standalone headings, commands, or key terms.

    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
    ﻭ
    0.78
    RS
    0.76
    Benchmarks
    0.68
    COMB
    0.67
    ﻤ
    0.66
    Hasil
    0.64
    hamed
    0.64
    stractions
    0.64
    jogo
    0.63
     руковод
    0.63
    POSITIVE LOGITS
    謊
    0.88
     flies
    0.86
     flagship
    0.86
     manganese
    0.85
     vacu
    0.84
     termite
    0.84
     buts
    0.84
     aunts
    0.81
     shea
    0.80
     alluvial
    0.80
    Activations Density 0.000%

    No Known Activations