© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-2B
    5. 24-GEMMASCOPE-TRANSCODER-16K
    7. 9683
    Prev
    Next
    INDEX
    Explanations

    unclear what this is finding, but I'd start by guessing it is finding specific nouns or proper names.

    oai_token-act-pair · gemini-2.0-flash

    Scientific/research text

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2b-pt-transcoders/layer_24/width_16k/average_l0_37
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.24.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
    ])));
    -0.57
    '}>
    -0.57
    ']>;
    -0.54
    .)}
    -0.53
     therefrom
    -0.52
    bbero
    -0.51
     Désolé
    -0.50
     »)
    -0.50
    viewDidLoad
    -0.49
    ]';
    -0.49
    POSITIVE LOGITS
    ftagPool
    0.67
    thâu
    0.60
     nhiêu
    0.57
    RegistryLite
    0.55
    Становништво
    0.54
     pinulongan
    0.53
     was
    0.52
    InitStruct
    0.52
     might
    0.49
     ویکی‌پدی
    0.48
    Activations Density 1.561%

    No Known Activations