© 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. 10-GEMMASCOPE-RES-131K
    7. 14636
    Prev
    Next
    INDEX
    Explanations

    once followed by an event

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

    The neuron is triggered by first‐person narrative markers—especially the pronoun “I” (and related “Once I” constructions).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
    enada
    -1.66
     Effekt
    -1.47
     divulgação
    -1.46
     sisält
    -1.37
    自从
    -1.35
     eszköz
    -1.34
     méret
    -1.32
     combinação
    -1.27
     transmissão
    -1.26
     domés
    -1.25
    POSITIVE LOGITS
     judicia
    1.57
     sowie
    1.41
     rende
    1.34
     например
    1.31
     "
    1.29
    社会的
    1.23
     Misalnya
    1.23
    IL
    1.20
    小伙伴
    1.20
     ruft
    1.19
    Activations Density 0.014%

    No Known Activations