© 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. 31810
    Prev
    Next
    INDEX
    Explanations

    lteconsulting

    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
    ness
    -0.28
    nya
    -0.26
    ss
    -0.24
    sWith
    -0.24
    ãģĦãģ¾ãģĻ
    -0.23
    set
    -0.22
    ย
    -0.21
    sh
    -0.21
    name
    -0.20
    sss
    -0.20
    POSITIVE LOGITS
    lement
    0.18
    ÑģÑĮ
    0.18
    urope
    0.18
    eeee
    0.17
    lements
    0.17
    lectric
    0.17
    ños
    0.16
    ãĥ¼ãĤ·ãĥ§ãĥ³
    0.16
    ño
    0.15
    eing
    0.15
    Activations Density 2.338%

    No Known Activations