© 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. 10-GEMMASCOPE-RES-131K
    7. 41477
    Prev
    Next
    INDEX
    Explanations

    analyze, add, stop, happened, believe, vote

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

    The neuron fires on first‐person pronouns and their accompanying auxiliaries (e.g. “I,” “we,” “can,” “let’s,” “’ll”), i.e. on self-referential or instructional speech.

    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
    Hoje
    -1.31
    Saiba
    -1.23
    Today
    -1.19
     acrescent
    -1.18
    彣
    -1.17
     ANCHE
    -1.12
     correto
    -1.10
     quaisquer
    -1.10
     OGSÅ
    -1.10
    Before
    -1.10
    POSITIVE LOGITS
     then
    1.38
    馳走
    1.16
     it
    1.09
    formes
    1.06
    给他
    1.05
    gridColumn
    1.05
    ificaciones
    1.04
     klat
    1.03
     sometimes
    1.02
     están
    1.02
    Activations Density 0.111%

    No Known Activations