© 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. 129125
    Prev
    Next
    INDEX
    Explanations

    The neuron activates on the end‐of‐block-comment delimiter (“*/”), marking the close of multi-line comments.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    programming documentation comments

    np_acts-logits-general · gemini-2.5-flash-lite
    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
    kesha
    -1.18
     concomitant
    -0.99
     FOUR
    -0.95
    很快就
    -0.93
    ков
    -0.93
    莫名
    -0.91
     других
    -0.91
    ladelphia
    -0.90
    偷偷
    -0.90
    ちら
    -0.89
    POSITIVE LOGITS
     its
    1.59
     every
    1.37
     at
    1.22
     which
    1.17
     their
    1.16
     now
    1.15
    Зміст
    1.09
    zji
    1.06
    História
    1.01
     begins
    1.00
    Activations Density 0.008%

    No Known Activations