© 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. 34-GEMMASCOPE-RES-131K
    7. 30720
    Prev
    Next
    INDEX
    Explanations

    weeks ago

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

    The neuron fires on expressions of time spans (e.g., “a few days,” “weeks ago”).

    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
     gebouw
    -1.04
    zial
    -1.00
    元々
    -0.99
     funzion
    -0.97
    ۂ
    -0.96
    vají
    -0.96
    ってました
    -0.96
    ponge
    -0.94
     كورة
    -0.94
     velikost
    -0.93
    POSITIVE LOGITS
     dozen
    1.38
    dozen
    1.38
     ago
    1.34
    <bos>
    1.10
     kilka
    1.02
    几
    0.98
     több
    0.97
     several
    0.97
    Several
    0.96
     hundred
    0.94
    Activations Density 0.035%

    No Known Activations