© 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. 33514
    Prev
    Next
    INDEX
    Explanations

    right thing

    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
    hips
    -0.10
    æĻ°
    -0.10
    ÑģÑĤÑĭ
    -0.10
    aida
    -0.10
    æ®Ĩ
    -0.09
    رات
    -0.09
    arket
    -0.09
    ราย
    -0.09
    aget
    -0.09
     slight
    -0.09
    POSITIVE LOGITS
    eous
    0.18
    zeitig
    0.15
    /left
    0.13
    -wing
    0.13
    most
    0.13
    iable
    0.13
    ness
    0.13
    -handed
    0.13
    -hand
    0.12
    -sized
    0.12
    Activations Density 0.068%

    No Known Activations