© 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. 95526
    Prev
    Next
    INDEX
    Explanations

    Evaluation of comments and code

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

    The neuron flags code tokens involved in runtime checks—assertions or `require` calls and logical‐comparison conditions.

    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
     any
    -1.30
     its
    -1.22
     they
    -1.13
    حول
    -0.95
     their
    -0.93
     все
    -0.91
     akzep
    -0.91
    ≳
    -0.91
     уже
    -0.90
    tara
    -0.90
    POSITIVE LOGITS
     muchacho
    1.08
    Carex
    1.03
     Osbourne
    1.02
    mmation
    0.96
    冂
    0.94
    uated
    0.94
     времена
    0.93
     سات
    0.92
    
    0.90
     tambin
    0.90
    Activations Density 0.000%

    No Known Activations