© 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. 19293
    Prev
    Next
    INDEX
    Explanations

    imaging technique

    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
    spiel
    -0.09
    stalk
    -0.08
    /autoload
    -0.08
    顺
    -0.08
    IPS
    -0.08
    pers
    -0.08
     post
    -0.08
    çĤ¸
    -0.08
    antine
    -0.08
    ellas
    -0.08
    POSITIVE LOGITS
    仪
    0.10
    ê°Ŀ
    0.10
    éĤ£ä½į
    0.09
    çĹķ
    0.09
    ÑĢин
    0.09
    angi
    0.09
    åħ¶æīĢ
    0.09
    è§ī
    0.09
    æĶ¿åºľéĩĩè´Ń
    0.09
    ruits
    0.09
    Activations Density 0.154%

    No Known Activations