© 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. 49198
    Prev
    Next
    INDEX
    Explanations

    numpy code blocks

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

    This neuron fires on tokens that are part of indented source‐code lines, i.e. code snippets rather than natural‐language text.

    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
     camisas
    -1.05
     penonton
    -1.05
     this
    -1.05
     it
    -1.02
    Have
    -1.00
    ఛ
    -0.99
     pembeli
    -0.99
     meneg
    -0.98
    Almost
    -0.96
    Still
    -0.95
    POSITIVE LOGITS
     unending
    0.91
     immense
    0.91
    くださった
    0.89
     kilka
    0.88
    一直在
    0.88
     fervent
    0.85
     allgeme
    0.85
    的情绪
    0.84
    fra
    0.84
    過了
    0.83
    Activations Density 0.004%

    No Known Activations