© 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
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 48468
    Prev
    Next
    INDEX
    Explanations

    I followed by a verb

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

    The neuron strongly activates on instances of the first‐person “I” (especially in direct speech in Biblical verses).

    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
     our
    -1.10
     orquí
    -0.96
     his
    -0.96
    我们是
    -0.88
    シャン
    -0.86
    讓我們
    -0.84
    letek
    -0.83
     we
    -0.83
    ıyors
    -0.83
    melhor
    -0.82
    POSITIVE LOGITS
     myself
    1.23
     Myself
    1.00
     will
    0.98
     I
    0.96
     your
    0.93
    ὴν
    0.87
     joaat
    0.86
     REQUIRED
    0.83
    ický
    0.82
    myself
    0.81
    Activations Density 0.009%

    No Known Activations