© 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. 27218
    Prev
    Next
    INDEX
    Explanations

    references and counts

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

    The neuron fires on code tokens related to references and reference‐counting (e.g. “ref,” “reference,” “weak,” “counted,” etc.).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    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
    Affirmed
    -0.90
    か
    -0.85
     ممكن
    -0.85
     hvert
    -0.84
     meisje
    -0.82
     sekä
    -0.80
    //
    -0.79
    àn
    -0.77
    грі
    -0.77
    ів
    -0.76
    POSITIVE LOGITS
    ագրություններ
    1.20
     its
    1.09
    教材
    1.04
    writeInt
    1.02
    DEPTH
    1.00
    何时
    0.97
    qPCR
    0.96
     CAROLINA
    0.95
    ﮧ
    0.95
     PETITION
    0.94
    Activations Density 0.005%

    No Known Activations