© 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. 10-GEMMASCOPE-RES-131K
    7. 45016
    Prev
    Next
    INDEX
    Explanations

    programming language names and code structures

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

    This neuron activates on forum‐style post metadata (e.g. “Originally Posted by,” “Posted by,” timestamps, usernames).

    oai_token-act-pair · o4-miniTriggered 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
    Not in Any Lists

    No Comments

    Negative Logits
     indruk
    -1.47
    臜
    -1.44
    的说道
    -1.41
     raiſ
    -1.41
    地说道
    -1.39
     gogh
    -1.38
    #
    -1.35
    liters
    -1.33
     versão
    -1.30
    웡
    -1.30
    POSITIVE LOGITS
    ⃙
    1.70
     mingle
    1.63
    都有
    1.60
    Hanya
    1.56
     but
    1.53
     which
    1.49
     kuiten
    1.46
     their
    1.42
    cją
    1.42
    ısından
    1.41
    Activations Density 0.004%

    No Known Activations