© 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. 128497
    Prev
    Next
    INDEX
    Explanations

    The neuron activates on words and symbols that denote appending or adding content (e.g. “append,” “+=,” “>>,” “add,” “appended”).

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

    add or append to existing

    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
    くれました
    -0.89
     wysokiej
    -0.88
    れています
    -0.86
    alumnos
    -0.85
    cuando
    -0.84
    事了
    -0.84
     LOST
    -0.83
     temprano
    -0.83
     całej
    -0.82
     نوم
    -0.82
    POSITIVE LOGITS
     additions
    1.31
     add
    1.27
     +=
    1.27
     adding
    1.27
     new
    1.18
     Adding
    1.17
     adds
    1.16
     доба
    1.15
     already
    1.11
     thêm
    1.07
    Activations Density 0.028%

    No Known Activations