© 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. 12061
    Prev
    Next
    INDEX
    Explanations

    we have

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

    The neuron detects first-person plural references, especially the pronoun "we" (and its contractions).

    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
     trimester
    -0.79
    络
    -0.77
    zan
    -0.77
    یده
    -0.77
    ocidad
    -0.76
    ush
    -0.75
     нака
    -0.74
     محفوظة
    -0.73
     asper
    -0.73
     Caritas
    -0.71
    POSITIVE LOGITS
     have
    1.59
     έχ
    1.25
     tenemos
    1.17
     we
    1.11
    有
    1.08
    Have
    1.02
     имеет
    0.91
     has
    0.91
     temos
    0.89
     máme
    0.85
    Activations Density 0.002%

    No Known Activations