© 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. 34-GEMMASCOPE-RES-131K
    7. 2227
    Prev
    Next
    INDEX
    Explanations

    dreams and nightmares

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

    The neuron is picking out mentions of “dream” (in any form: dream, dreams, titles or headings involving Dream) to identify dream-related content.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
     implacable
    -0.87
     hale
    -0.86
    RNA
    -0.86
     immu
    -0.84
    versorgung
    -0.81
    illa
    -0.80
    جلس
    -0.80
    ⊞
    -0.79
    uge
    -0.78
     intoxic
    -0.77
    POSITIVE LOGITS
     dream
    3.39
     dreams
    3.20
     dreaming
    2.77
    dream
    2.75
     Dream
    2.67
    Dreams
    2.58
     Dreams
    2.56
    dreams
    2.50
     dreamt
    2.44
     DREAM
    2.39
    Activations Density 0.034%

    No Known Activations