© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Gemma-2-27B
    3. 34-GEMMASCOPE-RES-131K
    4. 76001
    Prev
    Next
    INDEX
    Explanations

    `managed` `service` `identification` `property` `License`

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

    The neuron fires on software‐related metadata markers—things like version/build numbers, specification names, license terms, and other technical identifiers.

    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
     Briefe
    -0.78
    ocd
    -0.74
     デニム
    -0.74
    濺
    -0.74
     quitted
    -0.73
     заключается
    -0.73
     Jio
    -0.73
     名前
    -0.70
    νος
    -0.70
     開始
    -0.69
    POSITIVE LOGITS
     cuad
    0.71
     hos
    0.69
    攻撃
    0.65
    чик
    0.64
     sof
    0.63
     critique
    0.63
     muk
    0.63
    సు
    0.63
     obat
    0.61
    阻
    0.61
    Activations Density 0.057%

    No Known Activations