© 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. 3098
    Prev
    Next
    INDEX
    Explanations

    configuration setup

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

    The neuron fires on the quote‐mark tokens that open or close string literals in code (e.g. the " character delimiting JSON/JS/HTML strings).

    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
     célè
    -1.52
     montrant
    -1.47
    媖
    -1.34
    ほとん
    -1.32
     meub
    -1.30
     estadounid
    -1.28
     bascul
    -1.28
    presumably
    -1.28
     idiota
    -1.28
     inond
    -1.27
    POSITIVE LOGITS
     use
    1.54
     will
    1.31
     do
    1.20
     J
    1.20
     take
    1.20
     only
    1.19
    </em>
    1.16
     R
    1.16
     D
    1.16
    !!!!
    1.15
    Activations Density 0.054%

    No Known Activations