© 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. 84877
    Prev
    Next
    INDEX
    Explanations

    various forms of "rho"

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

    The neuron detects occurrences of the Greek‐letter name “rho” (i.e. any token “rho”/“Rho” in the text).

    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
    ?”
    -3.44
    ){
    -3.13
    诹
    -2.73
    }=\
    -2.67
    慜
    -2.56
    idjan
    -2.48
    
    -2.47
     μια
    -2.39
    ⎆
    -2.33
    𝕖
    -2.31
    POSITIVE LOGITS
    2
    2.92
    鱨
    2.73
    0
    2.48
    er
    2.30
    ’
    2.17
    {
    2.16
     sosial
    2.13
     minggu
    2.13
     núm
    2.11
    9
    2.08
    Activations Density 0.002%

    No Known Activations