© 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. 58782
    Prev
    Next
    INDEX
    Explanations

    various stages, open designs, of course

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

    The neuron fires on the very first content word of a new section or heading (i.e. the leading word at the start of a segment).

    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
    被他
    -1.60
    ...
    -1.57
     '
    -1.56
     sorprend
    -1.49
     nowego
    -1.43
    "
    -1.41
     disfrutar
    -1.40
     ausger
    -1.37
    ⍽
    -1.35
    それを
    -1.34
    POSITIVE LOGITS
    Adapun
    1.65
    鏵
    1.61
    Πηγή
    1.48
    ’
    1.48
    Berikut
    1.47
     ткани
    1.47
    Kemudian
    1.45
     catalana
    1.44
    Kamu
    1.41
    Berdasarkan
    1.39
    Activations Density 0.002%

    No Known Activations