© 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
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 119582
    Prev
    Next
    INDEX
    Explanations

    Judson, judo, glutathione

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

    the word "jud" at the beginning of words like "judge," "judicial," or "Judds."

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi

    The neuron activates on occurrences of “glutathione” (including its forms like “reduced glutathione” and the abbreviation “GSH”).

    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
    a
    -4.06
    i
    -3.84
    3
    -3.81
    1
    -3.77
    6
    -3.67
    8
    -3.48
    4
    -3.36
    7
    -3.31
    9
    -3.11
    5
    -3.05
    POSITIVE LOGITS
    罅
    3.00
     sorprendente
    2.95
     complicada
    2.73
    讴
    2.67
    嬗
    2.59
    輓
    2.45
    苹
    2.44
    咐
    2.42
     If
    2.41
    九年
    2.38
    Activations Density 0.020%

    No Known Activations