© 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. 15658
    Prev
    Next
    INDEX
    Explanations

    time, hour, month, year statistics

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

    The neuron fires on numeric literals (tokens composed of digits).

    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
    ymbol
    -1.36
     öf
    -1.21
    uransi
    -1.17
    tainable
    -1.15
    arture
    -1.15
    rophi
    -1.14
    ributes
    -1.13
    /******/
    -1.10
    datable
    -1.09
     fourrure
    -1.06
    POSITIVE LOGITS
     statistics
    1.31
     overall
    1.27
     year
    1.25
     each
    1.19
     month
    1.14
     trends
    1.13
     perguntou
    1.11
     statistical
    1.08
    别的
    1.03
    んですか
    1.02
    Activations Density 0.019%

    No Known Activations