© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 99121
    Prev
    Next
    INDEX
    Explanations

    duration of events

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

    The neuron fires on mentions of time spans or durations, especially numeric expressions of how long something lasts (e.g. “stop … lasts … seconds,” “lasted 45 minutes,” “four hours,” etc.).

    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
     薬
    -0.85
     ortaya
    -0.83
     groupings
    -0.79
    Jahre
    -0.78
     thyme
    -0.78
    Age
    -0.74
    \},
    -0.73
    mmas
    -0.73
    BOURNE
    -0.72
     lavoratori
    -0.72
    POSITIVE LOGITS
     lasted
    4.91
     lasts
    4.56
     lasting
    4.25
     duration
    3.59
    lasting
    3.20
     Duration
    2.91
     Lasting
    2.89
    duration
    2.88
    Duration
    2.81
     дли
    2.63
    Activations Density 0.099%

    No Known Activations