© 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. 63326
    Prev
    Next
    INDEX
    Explanations

    experiencing things

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

    The neuron strongly activates on personal‐reference words (e.g. “I,” “they,” “he”) and other pronouns that point to specific participants in the text.

    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
    AdapterView
    -0.84
    rpc
    -0.80
     lae
    -0.78
     estime
    -0.74
     Brugge
    -0.73
     הער
    -0.72
     hommage
    -0.71
     msgs
    -0.71
    оте
    -0.71
    lamó
    -0.69
    POSITIVE LOGITS
     experience
    6.06
     experiencing
    5.47
     experienced
    4.84
    experience
    4.78
     experiences
    4.75
     Experience
    4.47
    Experience
    4.41
     experiencia
    3.94
     EXPERIENCE
    3.83
     experien
    3.78
    Activations Density 0.100%

    No Known Activations