© 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. 1035
    Prev
    Next
    INDEX
    Explanations

    Trileaflet aortic valve reconstruction, evolution

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

    code structure in programming languages, particularly method/function declarations and class definitions.

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi

    The neuron fires strongly on identifiers and tokens involved in “bootstrapping” or setup—e.g. import lines pulling in config, `__init__` definitions, and other initialization-related names—so it’s looking for configuration/initialization 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
    pared
    -0.95
     животных
    -0.81
    boys
    -0.81
     Utf
    -0.79
    🌎
    -0.78
     StatusCode
    -0.76
    layed
    -0.76
    狞
    -0.76
    graphie
    -0.75
    €”
    -0.75
    POSITIVE LOGITS
    粽
    0.87
     then
    0.86
     either
    0.84
     whether
    0.83
    こちら
    0.81
    sonaro
    0.79
     such
    0.79
    covia
    0.74
     監督
    0.74
     muốn
    0.73
    Activations Density 0.003%

    No Known Activations