© 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-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 8661
    Prev
    Next
    INDEX
    Explanations

    way of saying

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

    The neuron is specifically detecting occurrences of the word “way” (e.g. in “way of,” “way to,” “another way,” etc.).

    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_31_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
     পদক্ষেপ
    0.63
    웰
    0.62
    ॠ
    0.62
     widely
    0.59
     Ос
    0.59
     esche
    0.59
     Adopt
    0.58
     grantees
    0.58
    </i>
    0.56
     estrategias
    0.56
    POSITIVE LOGITS
    ward
    0.82
     Way
    0.78
    way
    0.77
    yyyyyyyy
    0.75
    Way
    0.75
    yyyy
    0.66
    Ways
    0.66
    yy
    0.65
    ين
    0.65
     Ways
    0.65
    Activations Density 0.164%

    No Known Activations