© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 93407
    Prev
    Next
    INDEX
    Explanations

    stepping out of comfort zone

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

    The neuron fires on words that express personal judgment or evaluation—opinion-laden terms about belief, comfort, fear, “goodness,” 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_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
     réellement
    -1.03
    ικός
    -1.02
     dichter
    -1.02
    ჱ
    -1.01
     princípios
    -1.01
     attente
    -0.99
     QUALITY
    -0.96
     siè
    -0.96
     qa
    -0.95
    뗄
    -0.95
    POSITIVE LOGITS
     old
    1.46
     change
    1.38
     existing
    1.23
     habits
    1.23
     new
    1.21
     established
    1.13
    เดิม
    1.10
     
    1.09
     rigid
    1.09
     or
    1.09
    Activations Density 0.035%

    No Known Activations