© 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. 53802
    Prev
    Next
    INDEX
    Explanations

    software license and legal citation

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

    The neuron fires on the auxiliaries and past‐participles that mark passive‐voice constructions (e.g. “are,” “was,” “licensed,” “generated,” “unit,” “file,” “article” as they occur in passives).

    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
     this
    -0.93
     akku
    -0.90
    indexPath
    -0.90
    tldr
    -0.88
     komment
    -0.86
    quelize
    -0.85
    PUTY
    -0.83
    rectangle
    -0.82
    BITDA
    -0.82
    :[],
    -0.81
    POSITIVE LOGITS
    znego
    0.89
     is
    0.87
    igay
    0.82
     adipis
    0.81
    कि
    0.80
     joked
    0.80
    Sophie
    0.75
    ัส
    0.75
     radiological
    0.75
     पता
    0.73
    Activations Density 0.004%

    No Known Activations