© 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. 10967
    Prev
    Next
    INDEX
    Explanations

    ordinary or other categories

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

    The neuron fires on comparative or contrastive constructions (e.g. “more … than,” “like any other…”).

    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
     isla
    -0.95
     روند
    -0.94
    ophone
    -0.93
     produisons
    -0.89
    thodoxy
    -0.88
    儉
    -0.88
     prenda
    -0.87
    ardized
    -0.86
    styleable
    -0.85
     EXAMINATION
    -0.85
    POSITIVE LOGITS
     it
    1.12
     According
    0.96
     glut
    0.94
    кнул
    0.94
     PhD
    0.93
    Ｑ
    0.92
     These
    0.91
    𝑄
    0.91
    Here
    0.90
     Those
    0.90
    Activations Density 0.198%

    No Known Activations