© 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. 25687
    Prev
    Next
    INDEX
    Explanations

    units of measurement

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

    This neuron activates on tokens denoting CSS length units (e.g. “rem,” “em,” “dip,” “ex,” 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
    づけ
    -2.03
    にします
    -1.84
    の為
    -1.83
     and
    -1.82
    を行います
    -1.77
    をします
    -1.77
    んだよね
    -1.77
     goederen
    -1.74
    になりたい
    -1.74
     entendido
    -1.74
    POSITIVE LOGITS
    私は
    2.09
    これは
    1.98
    や
    1.96
    を
    1.95
    つまり
    1.93
    もちろん
    1.84
    では
    1.78
    iyaki
    1.77
    は
    1.77
    で
    1.74
    Activations Density 0.002%

    No Known Activations