© 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. 58531
    Prev
    Next
    INDEX
    Explanations

    description, code, cells

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

    The neuron fires on code‐style tokens—things like preprocessor marks (#), punctuation in identifiers (apostrophes, periods), digits, and file‐name or include syntax common in source code.

    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
     Inf
    -0.67
     ५
    -0.65
    diam
    -0.65
     foul
    -0.62
    Raja
    -0.62
     règle
    -0.61
     Fernsehen
    -0.61
    Ton
    -0.60
     Ton
    -0.60
     all
    -0.60
    POSITIVE LOGITS
     lts
    0.74
     Hohen
    0.72
     Manif
    0.71
    roz
    0.70
    äule
    0.70
     faim
    0.70
     reele
    0.69
     Amigos
    0.68
     quadru
    0.68
     vagu
    0.68
    Activations Density 0.044%

    No Known Activations