© 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. 6947
    Prev
    Next
    INDEX
    Explanations

    The neuron activates on modal auxiliary verbs and similar speculative or advisory constructions (e.g. could, might, should, don’t anticipate).

    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
     pourtant
    0.70
    ланган
    0.68
    િત
    0.66
    lands
    0.63
    因为
    0.61
    慧
    0.60
     révolution
    0.60
    瀚
    0.59
    比亚
    0.59
    alors
    0.59
    POSITIVE LOGITS
     Dostupné
    0.85
    쩔
    0.81
     Zahlen
    0.79
     understandably
    0.78
    oooo
    0.77
     естественно
    0.74
     necesariamente
    0.73
     chances
    0.71
     придется
    0.71
     indire
    0.70
    Activations Density 0.162%

    No Known Activations