© 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. 341
    Prev
    Next
    INDEX
    Explanations

    what I

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

    The neuron detects interrogative question words or phrases (e.g. what, why, how, are, will) used to form questions.

    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>
    0.38
    1
    0.37
    0.34
    2
    0.34
    0
    0.33
    .
    0.33
    )
    0.32
    }
    0.31
     as
    0.30
    :
    0.30
    POSITIVE LOGITS
    soever
    0.39
     તમે
    0.31
     Quelles
    0.30
    㢼
    0.30
     वाकई
    0.29
    jenigen
    0.29
     يمكن
    0.28
    λή
    0.28
    PARTMENT
    0.28
    ख्ती
    0.28
    Activations Density 0.936%

    No Known Activations