© 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. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 55339
    Prev
    Next
    INDEX
    Explanations

    page specific content

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    adamkarvonen/qwen3-32b-saes/saes_Qwen_Qwen3-32B_batch_top_k/resid_post_layer_32
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ships
    -0.13
    ship
    -0.12
    red
    -0.11
    shire
    -0.11
    reds
    -0.11
    uos
    -0.10
    PerPage
    -0.10
    shi
    -0.10
    åIJĦæł·
    -0.10
    ness
    -0.10
    POSITIVE LOGITS
    antry
    0.16
    åŃIJéĩĮ
    0.13
    peed
    0.12
    aus
    0.11
    alog
    0.11
    ableObject
    0.10
    ãģ¹ãģį
    0.10
    lets
    0.09
    enta
    0.09
    finding
    0.09
    Activations Density 0.076%

    No Known Activations