© 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. 18622
    Prev
    Next
    INDEX
    Explanations

    names like Hann, Hannibal, or ending in -ann

    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
    ../../../
    -0.13
    ا
    -0.11
    anos
    -0.11
    èĪ
    -0.10
    POSE
    -0.09
    ied
    -0.09
     Bundy
    -0.09
    æªĶ
    -0.09
    -wing
    -0.09
    most
    -0.09
    POSITIVE LOGITS
    ê¹IJ
    0.11
    ities
    0.11
    adir
    0.10
    æ§Į
    0.09
    istic
    0.09
    asan
    0.09
    ihilation
    0.08
    ÑĥÑĪки
    0.08
    alous
    0.08
    agram
    0.08
    Activations Density 0.055%

    No Known Activations