© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 6457
    Prev
    Next
    INDEX
    Explanations

    ", let"

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

    This neuron detects emphatic discourse markers and interjections (e.g. “Okay,” “Oh,” “Anyway,” “Yes,” “!”) that signal shifts or emphasis in tone.

    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
     oh
    1.03
    Oh
    0.99
    ,
    0.98
     Oh
    0.96
     โอ้
    0.92
    哦
    0.90
    Wow
    0.88
     哦
    0.83
    喔
    0.83
     Wow
    0.82
    POSITIVE LOGITS
     అలాంటి
    0.63
    evolving
    0.63
     விதி
    0.61
    そういう
    0.61
     அணுக
    0.59
     völlig
    0.57
     महत्त्वपूर्ण
    0.57
     মধ্যেও
    0.56
     veľ
    0.55
     அடிப்பட
    0.55
    Activations Density 0.174%

    No Known Activations