© 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. 53151
    Prev
    Next
    INDEX
    Explanations

    canal or canalisation

    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
    ãģ¹ãģį
    -0.10
    ening
    -0.10
    ãĥ³ãĤ°
    -0.10
    ened
    -0.09
    ầy
    -0.09
    esp
    -0.09
    çªģåĩºéĹ®é¢ĺ
    -0.09
    çŃĶåį·
    -0.08
    ä¸įé«ĺ
    -0.08
     Arch
    -0.08
    POSITIVE LOGITS
    icular
    0.11
    ëį°
    0.11
    çĪ·
    0.10
    RenderingContext
    0.10
    weed
    0.09
    绾
    0.09
    unes
    0.09
    unar
    0.09
    éķĴ
    0.09
    ÑģÑĤÑĭ
    0.09
    Activations Density 0.008%

    No Known Activations