© 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. 6212
    Prev
    Next
    INDEX
    Explanations

    mentions of the game Candy Crush and candy-related words.

    oai_token-act-pair · gemini-2.0-flash

    candy, toys, and fairy tales

    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
    شهاد
    -0.46
     detalhe
    -0.44
     un
    -0.42
    @@
    -0.42
     com
    -0.42
    erik
    -0.41
     mắc
    -0.41
    цом
    -0.41
    com
    -0.40
    平
    -0.40
    POSITIVE LOGITS
    tvguidetime
    0.92
     Houſe
    0.67
    MLLoader
    0.65
     doubtnut
    0.64
     rootReducer
    0.63
    IUrlHelper
    0.61
    niająca
    0.61
    TestingModule
    0.60
     Stande
    0.60
     houſe
    0.60
    Activations Density 2.742%

    No Known Activations