© 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. 11690
    Prev
    Next
    INDEX
    Explanations

    positive affirmations and praise

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

    The neuron strongly activates on enthusiastic praise and positive‐sentiment expressions (e.g. “beautiful,” “thank you,” “!”).

    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
    并非
    0.79
    并不是
    0.67
    如果我们
    0.67
    没办法
    0.65
     якщо
    0.65
     wrongfully
    0.64
     אם
    0.63
    だから
    0.63
    냐면
    0.63
     falsely
    0.61
    POSITIVE LOGITS
     Congratulations
    1.05
     congratulations
    1.03
     congrats
    1.03
     Congrats
    1.02
    Congratulations
    0.98
     admirable
    0.96
     impressive
    0.95
    Congrats
    0.91
     beeindruck
    0.90
     congrat
    0.88
    Activations Density 0.935%

    No Known Activations