© 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-2-2B
    5. 22-GEMMASCOPE-TRANSCODER-16K
    7. 3171
    Prev
    Next
    INDEX
    Explanations

    words or phrases relating to opera singers or music, especially French opera

    oai_token-act-pair · gemini-2.0-flash

    talent shows and music

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2b-pt-transcoders/layer_22/width_16k/average_l0_15
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.22.ln2.hook_normalized
    Architecture
    jumprelu_transcoder
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     propOrder
    -0.92
     ModelExpression
    -0.79
     doInBackground
    -0.76
    цездатний
    -0.75
     CreateTagHelper
    -0.74
     BoxFit
    -0.71
    PRNewswire
    -0.71
    CodeAttribute
    -0.70
    Билгалдахарш
    -0.68
     actor
    -0.67
    POSITIVE LOGITS
    πάρχ
    0.44
    Backing
    0.44
     talent
    0.44
     escola
    0.43
    reportError
    0.42
    廊下
    0.42
    الو
    0.41
    tal
    0.41
    sional
    0.41
     CASS
    0.41
    Activations Density 1.573%

    No Known Activations