© 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. 22-GEMMASCOPE-RES-131K
    7. 73089
    Prev
    Next
    INDEX
    Explanations

    quoted as saying

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

    The neuron detects words that report, cite, or attribute statements or sources (e.g., “quoted,” “cited,” “attributed,” “referred”).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
    谙
    -1.13
     grasp
    -1.03
     où
    -1.01
    中古
    -0.96
     contesta
    -0.92
    حياته
    -0.92
    葱
    -0.91
    щами
    -0.90
     rectifier
    -0.88
    年生
    -0.84
    POSITIVE LOGITS
     parecen
    1.06
    vlá
    1.05
     needing
    1.04
    REASONS
    1.04
    ldquo
    1.02
    voorbeeld
    1.02
     results
    1.00
     sarebbero
    1.00
     aloud
    1.00
    훤
    1.00
    Activations Density 0.024%

    No Known Activations