© 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. 22-GEMMASCOPE-RES-131K
    7. 35475
    Prev
    Next
    INDEX
    Explanations

    time points or durations

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

    The neuron responds to words that introduce temporal or conditional relations—terms like “when,” “immediately,” “before,” and “at.”

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
    expected
    -0.97
     ダブル
    -0.92
     herido
    -0.92
     pioggia
    -0.92
     femmin
    -0.91
    半
    -0.90
    Expected
    -0.90
    expects
    -0.90
     realme
    -0.89
    などが
    -0.89
    POSITIVE LOGITS
     at
    1.48
     after
    1.31
     time
    1.27
     September
    1.26
     their
    1.20
     January
    1.19
     before
    1.18
     December
    1.16
     start
    1.16
     when
    1.14
    Activations Density 0.053%

    No Known Activations