© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 45512
    Prev
    Next
    INDEX
    Explanations

    Actual values and parameters

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

    The neuron activates strongly on programming identifiers and other code‐specific symbols (e.g. class/type names, variable or function names, compiler flags, etc.).

    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
    𝚄
    -0.77
     Daytona
    -0.77
    zeuge
    -0.76
     स्त
    -0.74
    ონ
    -0.74
    珑
    -0.73
    ждение
    -0.73
    dfa
    -0.73
    Atlan
    -0.73
     Somali
    -0.72
    POSITIVE LOGITS
     actual
    0.86
    actual
    0.83
     Actual
    0.82
    ACTUAL
    0.81
     DESIGN
    0.78
     more
    0.76
     parameters
    0.75
    decision
    0.75
     Design
    0.73
    Actual
    0.73
    Activations Density 0.013%

    No Known Activations