© 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. 30436
    Prev
    Next
    INDEX
    Explanations

    data structure field terminators

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

    The neuron fires on standalone identifier‐style tokens—things like SQL keywords (e.g. CREATE, FIELDS), module/namespace names, and header/title words.

    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
     homic
    -0.80
     Anhang
    -0.75
    rute
    -0.73
     我的
    -0.72
     combines
    -0.72
     wśród
    -0.71
     στε
    -0.71
    cties
    -0.70
    ловая
    -0.70
    旋律
    -0.70
    POSITIVE LOGITS
     arkite
    0.96
    ایر
    0.93
     mem
    0.88
    Interview
    0.82
     LIQUID
    0.82
     たま
    0.81
    메
    0.81
    cuda
    0.80
    interview
    0.79
    beutel
    0.79
    Activations Density 0.001%

    No Known Activations