© 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. 54591
    Prev
    Next
    INDEX
    Explanations

    takes damage or harm

    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
    æŀľ
    -0.10
    åįıåķĨ
    -0.09
    ohn
    -0.09
     jo
    -0.09
    æĪ¡
    -0.08
     Lockheed
    -0.08
    æĶ»åħĭ
    -0.08
     Lup
    -0.08
    insky
    -0.07
    utherland
    -0.07
    POSITIVE LOGITS
    åıĹåĪ°äºĨ
    0.19
    åıĹåĪ°
    0.18
    éģŃåıĹ
    0.16
    éģŃåĪ°
    0.15
    éģŃ
    0.15
    åıĹ
    0.15
    æĶ¶åĪ°
    0.13
    ãĤĴåıĹãģij
    0.13
    åıĹãģij
    0.13
     recibir
    0.13
    Activations Density 0.042%

    No Known Activations