© 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. 34-GEMMASCOPE-RES-131K
    7. 64581
    Prev
    Next
    INDEX
    Explanations

    format and related structures

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

    The neuron strongly activates on words like “pattern(s),” “Pattern,” and related formatting tokens—i.e. it detects references to configuration patterns or format directives.

    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
    λους
    -0.82
     edades
    -0.77
     setTimeout
    -0.75
    setDuration
    -0.73
     jamás
    -0.71
    rhos
    -0.71
    빨
    -0.70
     blindness
    -0.69
     keyboard
    -0.69
    Moor
    -0.68
    POSITIVE LOGITS
     format
    1.50
    format
    1.48
    Format
    1.28
     formatting
    1.19
     formatter
    1.12
     Format
    1.10
     formats
    1.10
    FORMAT
    1.08
     формат
    1.07
    Formatter
    1.05
    Activations Density 0.071%

    No Known Activations