© 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. 12376
    Prev
    Next
    INDEX
    Explanations

    realization or surprise interjections

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

    This neuron detects informal conversational discourse markers and filler words typical of spoken-style text (e.g., “like,” “know,” “hey,” “great,” “thinking”).

    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
     Niall
    0.63
    ...@
    0.63
    codile
    0.62
     Ambulance
    0.59
    ？」
    0.59
    UCKY
    0.59
     प्लीज
    0.58
     قول
    0.58
     Lieutenant
    0.57
    UU
    0.57
    POSITIVE LOGITS
     hey
    1.02
     oh
    0.99
     Hey
    0.96
    Hey
    0.91
     Oh
    0.89
    Oh
    0.84
    噢
    0.81
     wow
    0.80
    Wow
    0.78
     Wow
    0.78
    Activations Density 0.120%

    No Known Activations