© 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-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 103174
    Prev
    Next
    INDEX
    Explanations

    it is possible to

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

    The neuron specializes in spotting the construction “possible to” (i.e. the word “possible” immediately followed by “to”).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     they
    -2.63
    >
    -1.87
    用
    -1.73
    没有
    -1.68
     identif
    -1.65
     specif
    -1.64
    感觉
    -1.63
    [];
    -1.61
    ILERS
    -1.60
    циями
    -1.60
    POSITIVE LOGITS
    烱
    1.81
     збіль
    1.73
    olescence
    1.70
     получает
    1.67
     нагрузки
    1.66
    棪
    1.60
     ascertain
    1.60
     завжди
    1.59
    epiece
    1.58
     зді
    1.57
    Activations Density 0.024%

    No Known Activations