© 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. 34-GEMMASCOPE-RES-131K
    4. 102907
    Prev
    Next
    INDEX
    Explanations

    coming to an end

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

    The neuron detects verbs and verb-phrases that signal something ending or reaching its conclusion (e.g. “coming to an end,” “came to an end,” “drawing … to a conclusion”).

    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
     recovered
    -0.78
    StringVar
    -0.75
     ゼ
    -0.75
    命运
    -0.72
     steigt
    -0.70
    alanya
    -0.68
     headshot
    -0.68
     أح
    -0.68
     figured
    -0.67
     بیرون
    -0.67
    POSITIVE LOGITS
     come
    2.16
     comes
    1.88
     came
    1.72
    Come
    1.66
    come
    1.57
    COME
    1.53
     coming
    1.48
     Come
    1.48
    comes
    1.48
    Came
    1.38
    Activations Density 0.024%

    No Known Activations