© 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. 0-GEMMASCOPE-TRANSCODER-16K
    7. 3680
    Prev
    Next
    INDEX
    Explanations

    words or partial words in Croatian related to Facebook, accounts, and email/SMS

    oai_token-act-pair · gemini-2.0-flash

    poš

    np_max-act · gemini-2.0-flash

    Croatian/Serbian

    np_max-act-logits · gpt-5-miniTriggered by @chenshw0109
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2b-pt-transcoders/layer_0/width_16k/average_l0_76
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.0.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
     hlavy
    -1.09
     vzduchu
    -1.04
     zvíř
    -1.01
     srdce
    -0.94
     tričko
    -0.93
    ově
    -0.93
     těž
    -0.92
     kří
    -0.91
     odpově
    -0.90
     ťaž
    -0.90
    POSITIVE LOGITS
     supérieurs
    0.96
     définiti
    0.91
     fermés
    0.90
     dévelo
    0.89
     hâte
    0.86
     réguli
    0.85
     fédé
    0.85
     supérieures
    0.85
     automatiques
    0.84
     chré
    0.84
    Activations Density 0.523%

    No Known Activations