© 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. 34-GEMMASCOPE-RES-131K
    7. 99584
    Prev
    Next
    INDEX
    Explanations

    mentions of usernames

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

    The neuron is most responsive to numeric tokens (digits and multi‐digit numbers).

    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
    》(
    -1.30
     trânsito
    -1.22
    喏
    -1.19
    場所に
    -1.09
    conector
    -1.09
    花费
    -1.05
    咳咳
    -1.05
    ologically
    -1.03
     nuestra
    -1.03
    場所で
    -1.02
    POSITIVE LOGITS
     thanks
    1.33
     what
    1.28
     because
    1.26
     thank
    1.25
     if
    1.23
     prêtre
    1.22
     why
    1.21
     exactly
    1.18
     yeah
    1.16
     если
    1.16
    Activations Density 0.006%

    No Known Activations