© 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. 70180
    Prev
    Next
    INDEX
    Explanations

    name or code identifier

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

    The neuron is a “code detector” – it lights up on programming‐language tokens (identifiers, keywords, punctuation, etc.) rather than on ordinary prose.

    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
     miras
    -0.86
     veau
    -0.84
     sete
    -0.83
     romano
    -0.82
     anos
    -0.82
     implica
    -0.81
    pylab
    -0.80
     vetor
    -0.79
     alcool
    -0.79
     paga
    -0.79
    POSITIVE LOGITS
     embodi
    0.87
    hythm
    0.86
     too
    0.85
    laboration
    0.85
    ほとんど
    0.82
    itatis
    0.81
    wuchs
    0.80
     respectivos
    0.79
    基本上
    0.78
    Moreover
    0.78
    Activations Density 0.003%

    No Known Activations