© 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-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 103925
    Prev
    Next
    INDEX
    Explanations

    top notch quality

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

    The neuron detects compound quality or excellence descriptors—especially hyphenated or paired modifiers signaling “high quality” or “top‐tier” (e.g. “top-notch,” “state-of-the-art,” “high-class,” “premier”).

    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
    屃
    -1.61
    琉
    -1.58
    𖤣
    -1.51
     schuldig
    -1.45
     permanentes
    -1.42
     geschenk
    -1.41
     reizen
    -1.41
     spørsmål
    -1.39
     สำหรับ
    -1.38
    "")
    -1.38
    POSITIVE LOGITS
    up
    1.74
    ">
    1.68
    If
    1.66
     quality
    1.65
    id
    1.52
    only
    1.49
    <strong>
    1.47
    AN
    1.47
    li
    1.45
    ab
    1.45
    Activations Density 0.028%

    No Known Activations