© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 3265
    Prev
    Next
    INDEX
    Explanations

    programming code and specific words

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

    The neuron fires on isolated label‐like tokens—single words or acronyms that stand alone as headings, column titles, macro names, or other section labels.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
    سات
    -0.68
     einzig
    -0.66
     problemer
    -0.66
     topo
    -0.63
    ICY
    -0.63
     pelear
    -0.62
     HEP
    -0.62
    目
    -0.62
    nossa
    -0.62
    COP
    -0.62
    POSITIVE LOGITS
    guarding
    0.77
     fennel
    0.73
     milla
    0.68
     Endocrinol
    0.68
     архивлан
    0.68
     Tangerang
    0.67
     виправивши
    0.66
     DeSantis
    0.65
    рија
    0.65
    founder
    0.65
    Activations Density 0.071%

    No Known Activations