© 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. 33029
    Prev
    Next
    INDEX
    Explanations

    just a phrase

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

    The neuron is responsive to the filler/hedging word “just” (and similar low-content adverbs used for emphasis or minimization).

    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
     بهذا
    -0.99
     этих
    -0.93
     limpeza
    -0.91
     więks
    -0.90
    uced
    -0.90
    Vw
    -0.90
    jil
    -0.89
     gän
    -0.88
     gutem
    -0.85
     suelos
    -0.85
    POSITIVE LOGITS
    ocidad
    0.93
    Both
    0.90
    сами
    0.89
    ctique
    0.87
     çok
    0.86
     Europas
    0.86
     área
    0.84
     häm
    0.83
     latter
    0.83
     sehr
    0.82
    Activations Density 0.067%

    No Known Activations