© 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. 14440
    Prev
    Next
    INDEX
    Explanations

    background and surrounding elements

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

    The neuron activates on technical descriptions of extracting or segmenting something “from” a background or noise context.

    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
     specialise
    -1.00
    諳
    -0.97
     unusual
    -0.97
     specialize
    -0.96
     three
    -0.95
    そこから
    -0.91
    ctility
    -0.91
     espectro
    -0.91
    tedrale
    -0.88
     disastrous
    -0.88
    POSITIVE LOGITS
     background
    1.10
     achtergrond
    1.06
    其他
    1.02
     surrounding
    1.02
    background
    1.02
    necessary
    0.98
     surround
    0.94
    totally
    0.93
     strona
    0.92
     ballast
    0.88
    Activations Density 0.134%

    No Known Activations