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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 103790
    Prev
    Next
    INDEX
    Explanations

    variant of

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

    The neuron activates on occurrences of the word “variant” (and its plural “variants”).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
    鷽
    -3.23
    蓐
    -3.02
     Diſ
    -2.94
     Inſ
    -2.89
    甞
    -2.88
    恹
    -2.84
    или
    -2.83
    鱨
    -2.78
    ꜆
    -2.77
    蔸
    -2.64
    POSITIVE LOGITS
     Being
    3.11
     Because
    3.11
     Having
    2.98
     Another
    2.88
     Since
    2.88
     After
    2.80
     Sometimes
    2.78
     Normally
    2.75
     Usually
    2.75
    During
    2.75
    Activations Density 0.006%

    No Known Activations