© 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. 31-GEMMASCOPE-2-RES-16K
    7. 10508
    Prev
    Next
    INDEX
    Explanations

    safety and comfort

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

    The neuron fires on phrases expressing gratitude for someone “taking the time,” i.e. occurrences of “take the time” (especially in “thanks for taking the time”).

    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
     Trouver
    0.64
     satisfactorily
    0.62
     verfügbar
    0.60
     conseguir
    0.56
     available
    0.55
     able
    0.55
     disponíveis
    0.53
    빈
    0.53
     coprime
    0.52
     succeed
    0.52
    POSITIVE LOGITS
     thanking
    0.71
    sız
    0.59
    吐槽
    0.57
     acknowledging
    0.56
     erlä
    0.56
     congratulated
    0.55
    ওর
    0.55
     explaining
    0.55
     thanked
    0.54
     اقول
    0.53
    Activations Density 0.099%

    No Known Activations