© 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. 15860
    Prev
    Next
    INDEX
    Explanations

    01:xx PM

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

    The neuron strongly activates on numeric tokens—especially dates, times, and other digit sequences.

    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
     December
    -1.07
     by
    -1.05
     November
    -1.05
     January
    -1.01
     dibu
    -0.97
     doesn
    -0.94
     two
    -0.92
     helped
    -0.91
     during
    -0.91
     another
    -0.91
    POSITIVE LOGITS
     nights
    1.42
    sleep
    1.20
     slumber
    1.17
     sleeping
    1.09
    ITTING
    1.08
     nachts
    1.08
     anderem
    1.03
    ̣c
    1.02
     promouvoir
    1.02
    Parcelize
    1.01
    Activations Density 0.014%

    No Known Activations