© 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. 116089
    Prev
    Next
    INDEX
    Explanations

    road bicycle | unexpected strict | court of

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

    This neuron fires strongly on the first word or symbol at the start of a new text block (e.g. the opening token of a section or paragraph).

    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
    ректор
    -0.91
    nière
    -0.79
    itre
    -0.79
     THEN
    -0.74
     three
    -0.72
     строительства
    -0.72
     rumor
    -0.71
    енера
    -0.71
     multitud
    -0.70
     moreno
    -0.70
    POSITIVE LOGITS
    mlung
    0.87
     bút
    0.81
    ⠀⠀
    0.80
    tawan
    0.79
    كلة
    0.77
    Obviously
    0.76
    と同じ
    0.76
    iklet
    0.75
     plait
    0.74
    とも
    0.74
    Activations Density 0.001%

    No Known Activations