© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 71903
    Prev
    Next
    INDEX
    Explanations

    tokens immediately preceding code snippets

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

    The neuron fires on programming‐style tokens—operators (like “=”, “+”), function calls, code keywords, and other syntax elements—i.e. it detects code snippets.

    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
    ئین
    -0.86
     endast
    -0.84
    ल्या
    -0.82
     nät
    -0.82
     atributos
    -0.82
     évek
    -0.82
     skydd
    -0.76
    ditions
    -0.75
     säl
    -0.75
     جديد
    -0.74
    POSITIVE LOGITS
     previous
    1.05
     earlier
    0.91
     adjacent
    0.88
    sidenav
    0.85
    jidad
    0.82
     previously
    0.81
     nearby
    0.80
    Earlier
    0.77
     ранее
    0.77
     ОТ
    0.77
    Activations Density 0.010%

    No Known Activations