© 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
    2. Gemma-3-12B
    3. 24-GEMMASCOPE-2-RES-16K
    4. 5481
    Prev
    Next
    INDEX
    Explanations

    idiomatic phrases

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

    The neuron responds strongly to familiar English idiomatic metaphor phrases (e.g. “bang for your buck,” “cherry on the cake,” “put your money where your mouth is”).

    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
    ارة
    0.70
    oreo
    0.70
    勂
    0.70
     કામ
    0.69
    লাপ
    0.67
    uttore
    0.67
    ixeira
    0.66
    য়ার
    0.66
    houette
    0.65
    रावट
    0.65
    POSITIVE LOGITS
     (
    0.75
     proverbial
    0.71
    …
    0.71
    ↵↵
    0.68
    (
    0.62
    (
    0.62
    ...
    0.60
    \...
    0.60
     nor
    0.57
     tých
    0.56
    Activations Density 0.019%

    No Known Activations