© 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. 34173
    Prev
    Next
    INDEX
    Explanations

    if then or consequences

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

    The neuron is primarily detecting numeric tokens—especially multi-digit numbers (including comma-separated thousands).

    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
     parfois
    -1.20
    sometimes
    -1.09
     autori
    -1.05
     corrom
    -1.05
     ieri
    -1.02
     gik
    -1.02
     様子
    -1.02
     souvent
    -1.01
     momente
    -0.99
    ربعة
    -0.98
    POSITIVE LOGITS
     will
    2.53
     would
    2.34
     then
    2.22
     sẽ
    1.47
     likely
    1.32
     I
    1.22
     будет
    1.21
     possible
    1.18
    Then
    1.17
    Would
    1.16
    Activations Density 0.016%

    No Known Activations