© 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. 10-GEMMASCOPE-RES-131K
    7. 7310
    Prev
    Next
    INDEX
    Explanations

    pp followed by parameters or code constructs

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

    The neuron fires on occurrences of the double-p sequence (“pp”) in tokens.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
    er
    -4.41
    ه
    -3.03
     provide
    -2.66
     To
    -2.38
    !”
    -2.33
    ”,
    -2.30
    They
    -2.28
    in
    -2.17
     
    -2.17
    ка
    -2.13
    POSITIVE LOGITS
    ússia
    2.59
    赑
    2.59
    ,.
    2.55
    瘩
    2.53
    鱨
    2.50
     firſt
    2.47
    꿩
    2.47
    簠
    2.44
    埽
    2.38
    捱
    2.34
    Activations Density 0.014%

    No Known Activations