© 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. 107737
    Prev
    Next
    INDEX
    Explanations

    time indicator am pm

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

    The neuron specializes in detecting publication timestamps (time‐of‐day markers such as “at 8:09 am,” “2:17 a.m.,” 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
    immediately
    -0.88
    ētu
    -0.87
    raid
    -0.84
    SCAPE
    -0.84
     (*(
    -0.83
    AspNetCore
    -0.81
    ेटा
    -0.79
     Cualquier
    -0.79
     *-
    -0.78
    chaft
    -0.76
    POSITIVE LOGITS
     reply
    0.85
     Reply
    0.83
    不仅
    0.79
    のち
    0.77
     hurt
    0.75
     lentes
    0.74
     respond
    0.73
     mention
    0.73
    这么说
    0.73
    tabli
    0.72
    Activations Density 0.009%

    No Known Activations