© 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
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 99313
    Prev
    Next
    INDEX
    Explanations

    state or condition of being

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

    The neuron fires on past-participial verb phrases (e.g. “could not be obtained,” “readily available,” “long been used,” “ramped up”).

    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
     $
    -1.41
    autoreleasepool
    -1.39
    )}$
    -1.37
     свои
    -1.32
    }(),
    -1.31
     (=
    -1.27
    %)$
    -1.26
    ニュアル
    -1.26
     ahora
    -1.25
     các
    -1.24
    POSITIVE LOGITS
     that
    1.56
    ↵↵
    1.44
     skóra
    1.34
     wesentlichen
    1.33
     particolar
    1.27
     haupts
    1.25
    なのだ
    1.25
     either
    1.25
     velocità
    1.23
     lichaam
    1.22
    Activations Density 0.412%

    No Known Activations