© 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. 5831
    Prev
    Next
    INDEX
    Explanations

    mentions of important topics and key points.

    oai_token-act-pair · gemini-2.0-flash

    A

    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
    most
    -0.72
    #+#
    -0.72
     most
    -0.72
    Most
    -0.66
     Most
    -0.65
    NOPQRST
    -0.63
    IVEREF
    -0.58
    UnusedPrivate
    -0.52
    ipedi
    -0.46
    Décès
    -0.46
    POSITIVE LOGITS
     Efq
    0.85
     Houſe
    0.82
     Jefus
    0.73
     Anſ
    0.71
     Shakspeare
    0.69
     Chriftian
    0.69
     Monfieur
    0.68
     Diſ
    0.68
     Perfon
    0.67
    SOUNDBITE
    0.67
    Activations Density 2.205%

    No Known Activations