© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 38629
    Prev
    Next
    INDEX
    Explanations

    `DisplayName` key-value programming codes

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

    The main thing this neuron does is detect numeric tokens (multi-digit numbers) 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_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
     sincron
    -0.86
     Einer
    -0.84
    Lastly
    -0.81
    ၄
    -0.80
     Приступљено
    -0.80
     قبل
    -0.78
    糠
    -0.78
     बिना
    -0.77
    Then
    -0.76
    kriv
    -0.76
    POSITIVE LOGITS
    łem
    0.78
     anytime
    0.77
    ături
    0.75
     美しい
    0.75
    avaş
    0.74
     gemeinsam
    0.74
     हैं
    0.74
     gefördert
    0.74
     entlang
    0.72
     Holistic
    0.72
    Activations Density 0.007%

    No Known Activations