© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 98784
    Prev
    Next
    INDEX
    Explanations

    java.nio.file.Files and strings

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

    The neuron fires on occurrences of well‐known library or utility names (e.g. Java std-lib classes like Objects, Files, Paths or Go packages like strings, errors).

    oai_token-act-pair · o4-mini
    Triggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Negative Logits
    だったら
    -1.78
    咆
    -1.66
     dann
    -1.61
    ve
    -1.58
     sinds
    -1.57
     Fünf
    -1.52
    磻
    -1.52
     freakin
    -1.51
    )
    -1.50
    珦
    -1.48
    POSITIVE LOGITS
    ズニー
    1.89
     ヶ
    1.85
     giebt
    1.81
    寕
    1.74
    .
    1.63
     汽車
    1.60
    尤其
    1.57
     cubano
    1.55
     conversations
    1.53
     douceur
    1.52
    Activations Density 0.003%

    No Known Activations