© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 66411
    Prev
    Next
    INDEX
    Explanations

    I statements

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

    The neuron is keyed to first-person singular self-references, i.e. “I,” “I’m,” “didn’t,” and other tokens in “I” statements.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
     sés
    -1.20
     elke
    -1.20
     merveilleux
    -1.19
    ÓW
    -1.13
    cobacterium
    -1.12
    ようになる
    -1.09
     betek
    -1.07
    tinh
    -1.06
    nadequate
    -1.06
    owaniu
    -1.06
    POSITIVE LOGITS
     will
    1.94
     would
    1.63
     want
    1.62
     not
    1.52
     cannot
    1.48
     does
    1.45
     only
    1.41
     know
    1.38
     neither
    1.38
     understand
    1.35
    Activations Density 0.017%

    No Known Activations