© 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. 1527
    Prev
    Next
    INDEX
    Explanations

    going off

    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
    å¹Ħ
    -0.11
    ãģĦãģĨ
    -0.10
    Ùģس
    -0.10
    olph
    -0.10
    oretical
    -0.10
    оÑģÑĮ
    -0.10
    izer
    -0.10
    åĽŀäºĭ
    -0.09
    èµ°åĩºæĿ¥
    -0.09
    ospel
    -0.09
    POSITIVE LOGITS
    ensively
    0.17
    /on
    0.13
    side
    0.13
    shoot
    0.12
    entimes
    0.12
    usion
    0.11
    eree
    0.11
    cial
    0.11
    /off
    0.10
    ICIAL
    0.10
    Activations Density 0.056%

    No Known Activations