© 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. 26491
    Prev
    Next
    INDEX
    Explanations

    foreign words and colors

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

    The neuron fires on typical figure‐caption language—graphical/statistical terms like “dot,” “line,” “axis,” “mean,” “confidence interval,” etc.—i.e. it detects text describing plots or images.

    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
     that
    -1.28
    嗰
    -1.11
     isn
    -1.09
    vestream
    -1.08
     According
    -1.05
     then
    -1.04
     even
    -1.02
    Ariana
    -1.02
     despite
    -1.01
     okay
    -1.01
    POSITIVE LOGITS
     място
    1.29
    ophanes
    1.16
     fatt
    1.11
     всеки
    1.09
     kiri
    1.09
     toko
    1.08
    过来的
    1.02
     represented
    1.02
     récemment
    1.02
     coloured
    1.01
    Activations Density 0.028%

    No Known Activations