© 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. 59190
    Prev
    Next
    INDEX
    Explanations

    feeling bad or dumb

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

    This neuron activates on first-person expressions of personal feelings or emotional states (e.g. “I feel…,” “I felt…,” plus self-descriptive terms like “stupid,” “dumb,” “safer,” 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
     hormon
    -0.96
     imposible
    -0.90
    (",",
    -0.88
    Со
    -0.87
     Amérique
    -0.83
     finan
    -0.82
     surpris
    -0.82
    และ
    -0.80
    henden
    -0.80
     amazement
    -0.79
    POSITIVE LOGITS
     bad
    2.63
    bad
    1.87
    Bad
    1.51
     Bad
    1.36
     silly
    1.30
     terrible
    1.27
     kötü
    1.24
    BAD
    1.23
     BAD
    1.20
     badly
    1.19
    Activations Density 0.014%

    No Known Activations