© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 84083
    Prev
    Next
    INDEX
    Explanations

    legal or court documents

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

    The neuron overwhelmingly activates on long numeric sequences or tokens—i.e. runs 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_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
     través
    -0.84
     selecting
    -0.83
    上野
    -0.82
    atleta
    -0.82
     large
    -0.81
     nahezu
    -0.81
    اضر
    -0.80
     first
    -0.80
     new
    -0.79
     four
    -0.79
    POSITIVE LOGITS
    utre
    1.05
    ModelSerializer
    0.86
     Llo
    0.86
    кзак
    0.85
    ћ
    0.83
    centering
    0.83
    isters
    0.83
    center
    0.82
    BatchNorm
    0.82
    whitespace
    0.82
    Activations Density 0.000%

    No Known Activations