© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 75890
    Prev
    Next
    INDEX
    Explanations

    specific numerical values

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

    The neuron fires on uppercase technical tokens—e.g. programming keywords, class/attribute names, or metadata labels (like DEFAULT, ARIA, PUBLIC)—i.e. code identifiers and protocol labels.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
    zov
    -0.73
    ัย
    -0.68
     funciona
    -0.68
    ffer
    -0.67
    aited
    -0.65
    想去
    -0.65
    thor
    -0.64
    Clicked
    -0.64
    bai
    -0.64
     builds
    -0.63
    POSITIVE LOGITS
    Vietnam
    0.76
    ীব
    0.73
    שם
    0.73
    Dragons
    0.72
    بره
    0.68
     Vietnam
    0.68
     ricevuto
    0.68
    imédia
    0.67
     Southland
    0.66
     ETI
    0.66
    Activations Density 0.060%

    No Known Activations