© 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. 4021
    Prev
    Next
    INDEX
    Explanations

    J followed by words or parts of words

    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
    ToJson
    -0.10
    Jwt
    -0.10
    JPEG
    -0.10
    æµļ
    -0.10
     jumlah
    -0.10
    éc
    -0.10
    ½
    -0.09
    ãĤ¸ãĥ§ãĥ³
    -0.09
    jective
    -0.09
    jpeg
    -0.09
    POSITIVE LOGITS
    uxtap
    0.15
    upyter
    0.15
    OURNAL
    0.15
    uggling
    0.14
    upiter
    0.14
     Bản
    0.13
    alous
    0.13
    aspers
    0.12
    apgolly
    0.12
    ansson
    0.12
    Activations Density 0.764%

    No Known Activations