© 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. 22-GEMMASCOPE-RES-131K
    7. 70492
    Prev
    Next
    INDEX
    Explanations

    My Tags

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

    The neuron activates on low‐level code listings—especially assembly/disassembly labels, directives, and long hexadecimal address or data sequences.

    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
     veda
    -1.14
     golos
    -1.06
     ziren
    -1.05
     metus
    -1.01
    neous
    -1.01
     kilom
    -1.00
     cinqu
    -0.96
     createSlice
    -0.94
     turista
    -0.94
     economia
    -0.93
    POSITIVE LOGITS
     but
    1.38
     at
    1.28
    agregar
    1.05
     during
    1.05
     dependiendo
    1.02
     titulada
    1.02
     portátiles
    1.02
    ヵ
    1.00
    ).
    0.99
     там
    0.99
    Activations Density 0.014%

    No Known Activations