© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 26341
    Prev
    Next
    INDEX
    Explanations

    duration or length of time

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

    The neuron is strongly triggered by longer, multi‐syllable words (e.g. “invalidating,” “official,” “assurances,” “receptors,” 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_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
    几种
    -0.79
    Attempts
    -0.79
     commer
    -0.77
    三种
    -0.77
    ピアノ
    -0.76
     biryani
    -0.76
     promulg
    -0.75
    裤
    -0.75
     plenti
    -0.74
     구성
    -0.74
    POSITIVE LOGITS
     during
    1.30
     time
    1.30
     gedurende
    1.20
     longer
    1.17
     day
    1.16
     long
    1.14
     late
    1.13
     lasting
    1.13
     overnight
    1.13
     протягом
    1.12
    Activations Density 0.273%

    No Known Activations