© 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. 47631
    Prev
    Next
    INDEX
    Explanations

    URL domains and code imports

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

    The neuron flags uncommon technical identifier tokens—things like code or markup names, IDs, class/file names, and other non-natural-language labels.

    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
     Recuperado
    -0.90
    CHORUS
    -0.82
    regards
    -0.82
    checkin
    -0.78
    androidx
    -0.78
     milder
    -0.77
    EAD
    -0.77
     }$
    -0.77
    şam
    -0.76
    verket
    -0.75
    POSITIVE LOGITS
    ,<
    0.85
     encaje
    0.80
    ?,
    0.80
     propagation
    0.78
     aceasta
    0.78
    지만
    0.77
    normal
    0.77
     sample
    0.77
     グリーン
    0.76
    up
    0.75
    Activations Density 0.000%

    No Known Activations