© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 21889
    Prev
    Next
    INDEX
    Explanations

    code generated

    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
    ively
    -0.10
    adier
    -0.10
    ss
    -0.09
    aya
    -0.09
    ned
    -0.09
    most
    -0.08
    loads
    -0.08
     Deadly
    -0.08
    ickers
    -0.08
    esse
    -0.08
    POSITIVE LOGITS
    pend
    0.14
    isplay
    0.12
    velop
    0.12
    umen
    0.11
    forces
    0.11
    0.11
    举æŃ¢
    0.11
    èļĤ
    0.10
    åĻĮ
    0.10
    mium
    0.10
    Activations Density 0.126%

    No Known Activations