© 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. 9855
    Prev
    Next
    INDEX
    Explanations

    handling `lot`, `interaction`, `completely`, `view`, `import`

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

    The neuron fires most strongly on discourse-level transition words or adverbials (e.g. Today, even, completely, just) that mark shifts or introductions of new statements.

    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
    <0x0D>
    0.88
     Senin
    0.80
     Trois
    0.79
     Wasn
    0.79
    鋭
    0.79
     Tricks
    0.77
    Hace
    0.76
    iyaki
    0.76
    Según
    0.76
    クリスマス
    0.75
    POSITIVE LOGITS
    м
    1.09
    ं
    0.95
    го
    0.93
    ции
    0.89
     дополни
    0.87
    ン
    0.86
    ם
    0.85
     cohom
    0.83
    сы
    0.82
     менее
    0.82
    Activations Density 0.000%

    No Known Activations