© 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. 84830
    Prev
    Next
    INDEX
    Explanations

    Mat or Anth prefixes

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

    The neuron activates strongly on occurrences of the letter sequence “anth” (as in ANTHONY or anthrax).

    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
    rogen
    -1.00
    Bilder
    -0.79
    eningen
    -0.79
    veland
    -0.77
     apostle
    -0.75
    ADDED
    -0.75
     Loyola
    -0.75
     передний
    -0.74
     шоколад
    -0.74
     pyrolysis
    -0.72
    POSITIVE LOGITS
    hew
    0.97
    thew
    0.77
    ew
    0.73
    hias
    0.71
    ws
    0.66
    delin
    0.65
    devise
    0.65
     Gästen
    0.64
     practised
    0.64
     Expect
    0.64
    Activations Density 0.041%

    No Known Activations