© 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. 4941
    Prev
    Next
    INDEX
    Explanations

    briefly

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

    The neuron fires on tokens that introduce or label feature lists and summary sections—e.g. section‐heading cues (“Sum ming Up:”, “features”, “amenities include”) and similar metadata/list indicators.

    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
    ástico
    0.82
     мужчины
    0.81
    ъем
    0.80
    ギリス
    0.79
    бой
    0.78
    лё
    0.76
    ភាព
    0.75
    ʈ
    0.75
     insides
    0.75
    мая
    0.74
    POSITIVE LOGITS
     lenne
    0.80
    h
    0.79
     chiama
    0.73
    WATER
    0.72
     적극
    0.71
     publique
    0.70
    ्ह
    0.69
     parce
    0.68
     ofere
    0.67
     esso
    0.67
    Activations Density 0.002%

    No Known Activations