© 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. 95340
    Prev
    Next
    INDEX
    Explanations

    firefox release dates

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

    The neuron primarily detects tokens that contain digits or apostrophes (i.e. numbers, dates, and possessive/contracted forms).

    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
    いって
    -0.99
    たびに
    -0.99
    作って
    -0.96
    arrête
    -0.96
    cillary
    -0.96
     vena
    -0.96
     vorbere
    -0.96
     Erkrank
    -0.96
     angebra
    -0.93
    Bö
    -0.93
    POSITIVE LOGITS
    ₆
    1.32
    ₃
    1.19
    ctober
    1.16
    IRQn
    1.12
    ₄
    1.12
     febru
    1.09
    umum
    1.08
    beda
    1.04
    Harga
    1.04
     Kompet
    1.02
    Activations Density 0.001%

    No Known Activations