© 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
    2. Gemma-2-27B
    3. 34-GEMMASCOPE-RES-131K
    4. 90980
    Prev
    Next
    INDEX
    Explanations

    total of

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

    The neuron strongly responds to tokens that appear in open-source license and copyright boilerplate (e.g. “forms,” “of,” “without,” “charge,” “AS IS”), i.e. it detects legal/license header language.

    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
     linda
    -0.85
     duran
    -0.83
    Cita
    -0.80
    Waktu
    -0.79
    żenia
    -0.79
     vaca
    -0.78
     both
    -0.77
     heartbreak
    -0.77
     other
    -0.77
     emot
    -0.77
    POSITIVE LOGITS
    ‚¬
    0.96
     cunos
    0.93
     among
    0.93
     friend
    0.91
     búsquedas
    0.90
    evre
    0.89
     acquaintances
    0.88
    itee
    0.87
     semelh
    0.86
    dillos
    0.86
    Activations Density 0.002%

    No Known Activations