© 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. 78887
    Prev
    Next
    INDEX
    Explanations

    list correspondence

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

    The main thing this neuron does is identify numerical expressions (e.g. counts, percentages, fold-changes) in the text.

    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
     suddenly
    -1.22
     even
    -1.08
     any
    -0.93
     that
    -0.93
     might
    -0.92
     sudden
    -0.92
     may
    -0.90
     its
    -0.90
     famous
    -0.87
     seldom
    -0.87
    POSITIVE LOGITS
     respectively
    9.19
    respectively
    7.13
     respectivamente
    6.56
     respectivement
    5.78
     соответственно
    5.38
     rispet
    4.69
     respective
    4.22
     respec
    4.19
    それぞれ
    4.16
    分别
    4.09
    Activations Density 0.274%

    No Known Activations