© 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
    2. Gemma-2-27B
    3. 22-GEMMASCOPE-RES-131K
    4. 66527
    Prev
    Next
    INDEX
    Explanations

    categories of things

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

    The neuron fires strongly on isolated, capitalized words that serve as section or heading titles.

    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
     triko
    -1.38
     atlet
    -1.28
     tribun
    -1.21
     anadolu
    -1.18
    preneurship
    -1.17
    ってください
    -1.16
    nitř
    -1.16
    NotFoundError
    -1.15
    hydrogenase
    -1.14
     antik
    -1.13
    POSITIVE LOGITS
     and
    1.49
     equipada
    1.12
     seguras
    1.09
     It
    1.05
     Not
    1.02
     Romains
    1.00
     Have
    0.98
     protegida
    0.98
    ándor
    0.98
    первых
    0.95
    Activations Density 0.034%

    No Known Activations