© 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. 21-GEMMASCOPE-TRANSCODER-16K
    7. 7881
    Prev
    Next
    INDEX
    Explanations

    words and phrases related to ancient history and mythology, particularly those involving battles and characters

    oai_token-act-pair · gemini-2.0-flash

    terms related to knights or chivalry

    np_token-act-pair-logits · gpt-4o-mini

    knights

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2b-pt-transcoders/layer_21/width_16k/average_l0_13
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.21.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
     Normdatei
    -0.69
    dersfield
    -0.63
    OGND
    -0.59
    </tfoot>
    -0.56
    كويكب
    -0.55
     okuyayım
    -0.54
    unhofer
    -0.53
     كورة
    -0.53
     umana
    -0.53
    ázaro
    -0.52
    POSITIVE LOGITS
    posedge
    0.65
    EDEFAULT
    0.63
    ScopeManager
    0.60
     Knights
    0.59
     chi
    0.59
    BagConstraints
    0.55
    LayoutStyle
    0.54
     knights
    0.53
     Knight
    0.52
    herald
    0.51
    Activations Density 1.646%

    No Known Activations