© 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. 39702
    Prev
    Next
    INDEX
    Explanations

    creating new objects with parameters

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

    The neuron flags actual code tokens—things like language keywords, identifiers, punctuation and literal constants—i.e. snippets of programming code rather than natural-language text.

    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
     including
    -0.98
     one
    -0.91
     with
    -0.88
     include
    -0.86
    including
    -0.84
     bậc
    -0.84
    with
    -0.83
    brow
    -0.83
     our
    -0.82
    톤
    -0.81
    POSITIVE LOGITS
     Envy
    0.94
    첫
    0.91
    Choi
    0.90
     бабу
    0.88
     tarefas
    0.87
    ингл
    0.87
     ДЕ
    0.86
     Мето
    0.85
    Castro
    0.84
    chsel
    0.84
    Activations Density 0.029%

    No Known Activations