© 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. 34-GEMMASCOPE-RES-131K
    7. 86875
    Prev
    Next
    INDEX
    Explanations

    using after nouns

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

    The neuron simply encodes each input token’s unique vocabulary index (i.e. it fires with a value equal to the token’s ID).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
    (!__
    -1.50
    ようになった
    -1.15
    ในการ
    -1.12
    ীয়
    -1.10
     ľah
    -1.09
     popularidad
    -1.02
    neder
    -1.02
    괌
    -0.98
     bateaux
    -0.98
    казыва
    -0.97
    POSITIVE LOGITS
     using
    8.69
    using
    5.53
     Using
    5.16
    Using
    5.16
     menggunakan
    5.09
     usando
    4.53
     USING
    4.28
     باستخدام
    4.16
     utilizando
    4.00
     используя
    3.95
    Activations Density 1.799%

    No Known Activations