© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 36103
    Prev
    Next
    INDEX
    Explanations

    Kavanaugh, Wiggins, wood

    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
    æĪ¾
    -0.11
    æ®ĭ
    -0.10
    人æŃ»äº¡
    -0.10
    ebb
    -0.10
    å¼Ģ设
    -0.09
    eron
    -0.09
    çĻĸ
    -0.09
    å¦ĤæľŁ
    -0.09
    rift
    -0.09
    大家ä¸Ģèµ·
    -0.08
    POSITIVE LOGITS
    ients
    0.10
    elli
    0.09
    DDS
    0.09
    è¶³å¤ŁçļĦ
    0.08
    imitive
    0.08
    æµģ失
    0.08
     Trusted
    0.08
    aval
    0.08
    仿
    0.08
    为空
    0.08
    Activations Density 0.008%

    No Known Activations