© 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. 22-GEMMASCOPE-RES-131K
    7. 25714
    Prev
    Next
    INDEX
    Explanations

    dataset or task names

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

    This neuron activates on code tokens that specify or name datasets and data‐loading configuration (e.g. dataset, data_path, loader, DataLoader).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
     how
    -1.35
     when
    -1.23
     an
    -1.17
     sometimes
    -1.14
    */
    -1.13
     this
    -1.12
     others
    -1.11
     often
    -1.11
    ;
    -1.09
     while
    -1.09
    POSITIVE LOGITS
     négo
    1.29
     défis
    1.23
     réserve
    1.22
     étapes
    1.21
    ̹
    1.17
     possibilités
    1.16
     côtes
    1.14
    Égypte
    1.13
    展现
    1.13
    erreur
    1.13
    Activations Density 0.015%

    No Known Activations