© 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-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 14176
    Prev
    Next
    INDEX
    Explanations

    that's followed by explanation or confirmation

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

    The neuron is keyed to sentence-level discourse markers—especially the “That’s …” construction (and related cue words like “because” or “right”) that kick off or link sentences.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     щодо
    -1.66
    這些
    -1.66
    其他
    -1.65
    是
    -1.63
     Completely
    -1.58
     джере
    -1.55
    这些
    -1.54
     Initially
    -1.53
     bekannten
    -1.53
     =
    -1.52
    POSITIVE LOGITS
     tradic
    1.68
     celebrid
    1.65
    ꪳ
    1.61
     incroyable
    1.60
     elegante
    1.59
    recently
    1.59
     induk
    1.58
     komer
    1.56
     italien
    1.55
     PATENT
    1.55
    Activations Density 0.019%

    No Known Activations