© 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. 61907
    Prev
    Next
    INDEX
    Explanations

    AM frequency tuning methods

    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
    è°ĭ
    -0.09
    è¾ĵ
    -0.08
    éķĽ
    -0.08
    æŃ¢
    -0.08
    .Companion
    -0.08
    Moder
    -0.08
    æ¯ıä½į
    -0.08
     Abuse
    -0.08
     Pay
    -0.08
    Pay
    -0.08
    POSITIVE LOGITS
    ORIA
    0.10
    ä¸ĢéŨ
    0.10
    åĵ¼
    0.09
    çĽijçĿ£ç®¡çIJĨ
    0.09
    对åħ¶è¿Ľè¡Į
    0.08
    çĨĻ
    0.08
    ÑĤик
    0.08
    -BEGIN
    0.08
    çĨ¹
    0.08
    orest
    0.08
    Activations Density 0.079%

    No Known Activations