© 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. 64776
    Prev
    Next
    INDEX
    Explanations

    listing items with "and"

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

    The neuron fires on numeric tokens (counts, measurements, numeric values, especially decimals).

    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
     HttpStatus
    -0.89
     würd
    -0.80
     対応
    -0.79
    timacy
    -0.76
    まずは
    -0.74
    はもちろん
    -0.72
     Qty
    -0.69
     girasol
    -0.69
     Akhtar
    -0.69
    次に
    -0.69
    POSITIVE LOGITS
     overall
    2.67
     combined
    2.30
    Overall
    1.99
    overall
    1.95
     Overall
    1.95
     combine
    1.84
     Combined
    1.80
    OVERALL
    1.79
    Combined
    1.79
    combined
    1.76
    Activations Density 0.063%

    No Known Activations