© 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. 20-GEMMASCOPE-TRANSCODER-16K
    7. 11773
    Prev
    Next
    INDEX
    Explanations

    words starting with "hyper" (or "hypo") and sometimes words with "cap"

    oai_token-act-pair · gemini-2.0-flash

    hyper

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2b-pt-transcoders/layer_20/width_16k/average_l0_11
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.20.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
     patrones
    -0.50
     članak
    -0.49
    وب
    -0.46
     CreateTagHelper
    -0.45
    RestController
    -0.44
    ]")]
    -0.44
    extérieur
    -0.43
    meanor
    -0.43
    baomidou
    -0.43
    블
    -0.42
    POSITIVE LOGITS
     houſe
    1.03
     Reſ
    0.96
     Houſe
    0.96
     whoſe
    0.95
     ſeveral
    0.94
     Jefus
    0.93
     myſelf
    0.92
     Theſe
    0.91
     itſelf
    0.91
     greateſt
    0.90
    Activations Density 0.595%

    No Known Activations