© 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. 22-GEMMASCOPE-RES-131K
    7. 15574
    Prev
    Next
    INDEX
    Explanations

    names and technical terms

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

    The neuron activates strongly on programming‐style syntax tokens—especially braces, indentation whitespace, and other code delimiters.

    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
     that
    -1.20
    💀
    -1.07
     autobus
    -1.05
    alaska
    -1.05
     savon
    -1.05
     acrylique
    -1.00
    КТ
    -0.99
    alors
    -0.99
    fontFamily
    -0.98
     materials
    -0.97
    POSITIVE LOGITS
     at
    0.91
     íntegra
    0.85
     them
    0.85
    過ご
    0.84
    vgili
    0.83
    同じく
    0.83
     request
    0.82
    他們
    0.82
     идея
    0.82
     北欧
    0.82
    Activations Density 0.000%

    No Known Activations