© 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. 10880
    Prev
    Next
    INDEX
    Explanations

    looks like it refers

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

    The neuron fires strongly on epistemic and copular verbs expressing belief or stance (e.g. thought, believe, was, be).

    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
    Change
    0.66
     requirement
    0.65
     രണ്ടു
    0.64
    Lorsque
    0.64
    对于
    0.63
    如何在
    0.63
    +=
    0.59
     change
    0.58
    Введите
    0.58
    對於
    0.58
    POSITIVE LOGITS
     origen
    0.81
     origine
    0.80
    真相
    0.74
     unmist
    0.71
     culprits
    0.71
     clues
    0.71
     culprit
    0.70
    実
    0.68
     pistas
    0.68
     sospe
    0.66
    Activations Density 0.204%

    No Known Activations