© 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. 7245
    Prev
    Next
    INDEX
    Explanations

    auxiliary verbs such as would, could, will, can, and should, often in association with personal pronouns

    oai_token-act-pair · gemini-2.0-flash

    contextual and procedural expressions

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

    will would

    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
    +#+#
    -0.67
    setupUi
    -0.64
    BURGH
    -0.60
    calaure
    -0.54
    hoi
    -0.54
     Addis
    -0.54
    фраструк
    -0.53
    cplusplus
    -0.52
     geprek
    -0.50
    preduce
    -0.50
    POSITIVE LOGITS
    BeginContext
    0.56
    RTEE
    0.54
    ValueStyle
    0.50
    AddHtmlAttribute
    0.49
    princ
    0.48
     Réponses
    0.48
    alami
    0.47
     StringTokenizer
    0.47
    ніципалі
    0.47
     nonUne
    0.46
    Activations Density 1.130%

    No Known Activations