© 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. 34-GEMMASCOPE-RES-131K
    7. 103200
    Prev
    Next
    INDEX
    Explanations

    feel good

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

    The neuron primarily activates on occurrences of the substring “feel” (as in feel, feel-good, feelers, 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_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
    óricas
    -0.74
     治療
    -0.74
     iny
    -0.72
    mäßige
    -0.71
     bestä
    -0.71
    щаться
    -0.70
    協力
    -0.69
     pokoje
    -0.69
     Amalia
    -0.69
     лимон
    -0.69
    POSITIVE LOGITS
     feel
    3.00
     Feel
    2.48
    feel
    2.42
    Feel
    2.39
     FEEL
    1.71
     feels
    1.39
    good
    1.37
    feels
    1.20
     good
    1.17
    ings
    1.13
    Activations Density 0.009%

    No Known Activations