© 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. 6094
    Prev
    Next
    INDEX
    Explanations

    the stems of multi-syllable words ending in "y", "r" or a vowel followed by "er"

    oai_token-act-pair · gemini-2.0-flash

    technical writings

    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
    ніципалі
    -0.50
     createContext
    -0.48
    ymce
    -0.44
    );?>
    -0.43
     auto
    -0.43
    usercontent
    -0.43
     становника
    -0.43
    KommentareTeilen
    -0.42
     cr
    -0.42
    routeProvider
    -0.42
    POSITIVE LOGITS
    LookAnd
    0.72
    IsContent
    0.71
    oa̍t
    0.66
     itſelf
    0.66
     Theſe
    0.65
    IonicModule
    0.65
     propOrder
    0.64
     kaarangay
    0.63
     Reſ
    0.62
     Houſe
    0.62
    Activations Density 1.255%

    No Known Activations