© 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. 130804
    Prev
    Next
    INDEX
    Explanations

    The neuron responds to words that convey subjective evaluation or degree—i.e. opinionated or judgment‐bearing terms (like “beautiful,” “phenomenal,” “sophisticated,” “too,” “all,” etc.).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    overall thoughts

    np_acts-logits-general · gemini-2.5-flash-lite
    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
    mogen
    -1.23
     taşlı
    -1.18
    ebly
    -1.06
    lysm
    -1.02
     này
    -1.02
    的是
    -1.02
     删除
    -1.02
     izle
    -1.02
    ighed
    -0.99
     هذا
    -0.96
    POSITIVE LOGITS
     the
    1.75
     overall
    1.45
    overall
    1.26
    Overall
    1.19
     likewise
    1.14
     there
    1.12
     Overall
    1.01
     ajudam
    0.98
    整體
    0.93
     you
    0.93
    Activations Density 0.023%

    No Known Activations