© 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
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 15086
    Prev
    Next
    INDEX
    Explanations

    not sure if

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

    The neuron strongly activates on the demonstrative pronoun “this” (especially in the phrase “this is”) and second-person “you” (as in “you have”), i.e. it detects those common introductory pronouns in questions or 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_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
     mesmerizing
    -2.05
    专业的
    -1.86
    饜
    -1.78
    を作って
    -1.77
    𓁹
    -1.77
    骉
    -1.76
    僰
    -1.73
    ſſen
    -1.72
    跩
    -1.69
    ladolid
    -1.66
    POSITIVE LOGITS
     any
    2.00
    </i>
    1.97
     using
    1.78
     chodzi
    1.76
     because
    1.75
     after
    1.71
     will
    1.71
     were
    1.70
     should
    1.64
     związane
    1.63
    Activations Density 0.039%

    No Known Activations