© 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. 64703
    Prev
    Next
    INDEX
    Explanations

    dates and titles

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

    The neuron is highly selective for the uppercase phrase “TO SHOW” (as in the legal heading “Order to Show Cause”).

    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
     without
    -1.76
     syk
    -1.43
     when
    -1.43
     conector
    -1.38
     Desember
    -1.37
     Septembre
    -1.36
     paco
    -1.36
    🌟
    -1.35
     qtd
    -1.35
     Kecil
    -1.34
    POSITIVE LOGITS
     myriad
    1.56
     crucial
    1.51
     energías
    1.50
     overly
    1.49
     praticamente
    1.48
     startling
    1.47
     dando
    1.46
     immense
    1.45
     diversos
    1.42
     distintos
    1.41
    Activations Density 0.001%

    No Known Activations