© 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. 15197
    Prev
    Next
    INDEX
    Explanations

    being impressed

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

    The neuron detects occurrences of the root “impress,” i.e. words like “impressed” or “impressive.”

    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
     a
    -2.16
     \&
    -2.06
    न
    -2.02
    N
    -1.95
                
    -1.94
    同伴
    -1.84
     \\
    -1.75
    ৫
    -1.75
     '
    -1.73
    .*")]
    -1.72
    POSITIVE LOGITS
     impresion
    1.99
    Surprisingly
    1.77
     daer
    1.73
     uye
    1.72
     wanda
    1.69
     kyo
    1.65
     romper
    1.63
    踅
    1.62
     olor
    1.61
    ﱠ
    1.60
    Activations Density 0.006%

    No Known Activations