© 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. 30882
    Prev
    Next
    INDEX
    Explanations

    python function calls and assignments

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

    This neuron activates on numeric constants in code, i.e. when integer (or other numeric) literals appear.

    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
     they
    -0.91
     their
    -0.83
    듬
    -0.76
    ี้ยง
    -0.76
     kecelakaan
    -0.75
     startDate
    -0.75
    往下
    -0.75
    )]:
    -0.74
     Argentina
    -0.74
     there
    -0.74
    POSITIVE LOGITS
     f
    2.20
    f
    1.70
     fumble
    0.94
    ($"
    0.94
     F
    0.90
    着脸
    0.88
    ={`${
    0.85
    OKA
    0.84
    ॉन
    0.82
     rostros
    0.82
    Activations Density 0.013%

    No Known Activations