© 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. 14428
    Prev
    Next
    INDEX
    Explanations

    <!-- and -->

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

    The neuron activates on programming‐language syntax tokens—like braces (“{”), attribute‐assignment markers (e.g. controller="…"), import/service keywords—i.e. code structural elements.

    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
     enkele
    -1.09
     Treff
    -1.01
     rosario
    -1.01
    кса
    -1.00
     afforded
    -0.99
    lin
    -0.98
     satria
    -0.98
    早已
    -0.96
     begegnen
    -0.95
     wickedness
    -0.94
    POSITIVE LOGITS
     Ahnung
    1.21
    你现在
    1.14
    urusan
    1.02
    GOTREF
    1.02
     натя
    1.02
    إذا
    1.00
     teren
    0.98
     tedy
    0.98
     []:
    0.95
    ผม
    0.95
    Activations Density 0.000%

    No Known Activations