© 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. 32270
    Prev
    Next
    INDEX
    Explanations

    harmful

    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
    chia
    -0.11
    verts
    -0.11
    ÑĥÑĩа
    -0.10
    ammen
    -0.10
    ubo
    -0.10
    owards
    -0.09
    UnderTest
    -0.09
     ÑĤеп
    -0.09
    itis
    -0.09
    owan
    -0.09
    POSITIVE LOGITS
    lessly
    0.18
    fully
    0.17
    ful
    0.14
    lessness
    0.12
    less
    0.11
    ware
    0.10
    FUL
    0.10
    PING
    0.10
    çijģ
    0.10
    å¦Ĥæľīä¾µæĿĥ
    0.09
    Activations Density 0.010%

    No Known Activations