© 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
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 70459
    Prev
    Next
    INDEX
    Explanations

    pronoun subject followed by verb

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

    This neuron detects instances of reported speech or dialogue attribution—especially sentences that begin with a pronoun (He/She) introducing what someone says or does, often marked by a colon.

    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
     them
    -1.85
    ła
    -1.75
     say
    -1.74
    那些
    -1.73
    mohon
    -1.70
     той
    -1.58
     slik
    -1.56
     och
    -1.55
    その
    -1.52
    льные
    -1.52
    POSITIVE LOGITS
    を紹介
    1.94
     poté
    1.88
    起码
    1.75
     Anſ
    1.72
     aseguró
    1.71
    ㈱
    1.70
    ネットで
    1.64
     vroeger
    1.63
     ofrecemos
    1.58
    他们是
    1.52
    Activations Density 0.022%

    No Known Activations