© 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. 10783
    Prev
    Next
    INDEX
    Explanations

    q followed by '.ajax' or 'engine'

    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
    posted
    -0.09
    ized
    -0.09
    stood
    -0.09
    åĪ«è¯´
    -0.09
    bery
    -0.09
    Absent
    -0.08
    ละ
    -0.08
    ange
    -0.08
    ads
    -0.08
    REA
    -0.08
    POSITIVE LOGITS
    sWith
    0.15
    ubits
    0.13
    ubit
    0.13
    uality
    0.12
    åĭĻ
    0.11
    0.10
    nowledge
    0.10
    plorer
    0.10
    itude
    0.10
    ademic
    0.10
    Activations Density 0.098%

    No Known Activations