© 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. 7511
    Prev
    Next
    INDEX
    Explanations

    words used in scientific papers, particularly those using models and discussing results with validation.

    oai_token-act-pair · gemini-2.0-flash

    obsolete or historical language forms

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

    scientific/experimental studies

    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
    awaiter
    -0.55
     transférez
    -0.48
     قول
    -0.47
     friendly
    -0.47
    ัพท์
    -0.47
    候
    -0.46
    ارت
    -0.45
     circo
    -0.44
     coinciden
    -0.44
    htdocs
    -0.43
    POSITIVE LOGITS
     ſever
    0.81
    ftagPool
    0.74
     purpoſe
    0.69
     Jefus
    0.68
    findpost
    0.67
     myſelf
    0.67
     pleaſure
    0.66
    ^(@)
    0.64
     ſeveral
    0.64
     uſ
    0.63
    Activations Density 16.559%

    No Known Activations