© 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. 953
    Prev
    Next
    INDEX
    Explanations

    Match DetailActivity

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

    The neuron specializes in spotting code‐style identifiers and keywords (e.g. import statements, class/package names, annotations, or other programming‐language tokens).

    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
     furt
    -0.95
     fiducia
    -0.82
    glGen
    -0.80
     assa
    -0.79
    FX
    -0.75
     akhirnya
    -0.75
     Cheddar
    -0.75
     desenho
    -0.75
    随意
    -0.74
     enunci
    -0.74
    POSITIVE LOGITS
     examine
    0.77
    entir
    0.74
    upsert
    0.73
     £
    0.72
    readAll
    0.71
     Media
    0.71
    涌
    0.71
     зрения
    0.71
     ordre
    0.70
     abundante
    0.70
    Activations Density 0.001%

    No Known Activations