© 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. 76203
    Prev
    Next
    INDEX
    Explanations

    encoding types

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

    The neuron fires strongly on “encoding” or “Encoding” fields in file headers or comments, i.e. lines that declare the text/file encoding.

    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
     often
    -0.90
     more
    -0.86
     while
    -0.84
     or
    -0.82
     found
    -0.79
    回数
    -0.78
     prominent
    -0.77
    Often
    -0.77
     discovered
    -0.76
     horizontally
    -0.76
    POSITIVE LOGITS
    application
    0.97
     majestic
    0.93
    seemed
    0.91
     textiles
    0.90
    probably
    0.90
     aplicación
    0.90
    TEX
    0.88
     jab
    0.88
    ourced
    0.87
    amssymb
    0.87
    Activations Density 0.001%

    No Known Activations