© 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. 44756
    Prev
    Next
    INDEX
    Explanations

    reporting speech or facts

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

    This neuron detects journalistic attribution phrases—especially reporting verbs like “says,” “said,” “reports,” etc.

    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
     selepas
    -1.53
    膂
    -1.39
     péché
    -1.32
     केवल
    -1.30
     bénéficiaire
    -1.28
    FICHA
    -1.27
     Siapa
    -1.27
     eerst
    -1.27
    новременно
    -1.25
    SOBRE
    -1.24
    POSITIVE LOGITS
     help
    2.08
     didn
    1.74
     aren
    1.63
     today
    1.58
     their
    1.57
     doesn
    1.56
     they
    1.54
     especially
    1.51
     don
    1.49
     isn
    1.48
    Activations Density 0.139%

    No Known Activations