© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 129676
    Prev
    Next
    INDEX
    Explanations

    Neuron 4 detects long runs of repeated punctuation characters (like dashes or equals signs) used as divider or separator lines in code comments.

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

    code comment separators

    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
     sés
    -1.51
     pétrole
    -1.46
    Peut
    -1.45
    媄
    -1.45
     fidé
    -1.44
     ſou
    -1.44
    鳚
    -1.44
    蕺
    -1.44
     autocollant
    -1.43
     miroir
    -1.42
    POSITIVE LOGITS
    j
    1.88
    z
    1.84
    R
    1.84
    G
    1.77
    k
    1.74
    N
    1.67
    d
    1.67
    H
    1.65
    C
    1.63
    O
    1.63
    Activations Density 0.032%

    No Known Activations