© 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. 7119
    Prev
    Next
    INDEX
    Explanations

    tt or TT followed by file

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

    The neuron is specialized to detect the specific uppercase bigram “TT” (i.e. two consecutive capital T’s).

    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
    halle
    -0.75
     Geno
    -0.73
     İç
    -0.70
     thoracic
    -0.70
     Bees
    -0.70
     Enter
    -0.69
     дат
    -0.69
     טע
    -0.68
     mito
    -0.68
     Epic
    -0.68
    POSITIVE LOGITS
    vrigt
    0.85
     podpis
    0.80
    icides
    0.80
    ksha
    0.75
     práctica
    0.73
    menea
    0.73
    癸
    0.73
    uanya
    0.71
     PUEDE
    0.71
    ometric
    0.70
    Activations Density 0.035%

    No Known Activations